正在加载视频...

视频加载失败

返回首页

[VAE] by Hand ✍️ A Variational Auto Encoder (VAE) learns the structure (mean and variance) of hidden features and generates new data from the learned structure. In contrast, GANs only learn to generate new data to fool a discriminator; they may not necessarily know the underlying structure of the...

ProfTomYeh's profile picture

Tom Yeh

65,666 subscribers

48,529 次观看 • 2 年前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Batch Normalization by hand ✍️ ~ 7 steps walkthrough below Batch normalization is common practice for improving training and achieving faster convergence. It sounds simple. But it is often misunderstood. 🤔 Does batch normalization involve trainable parameters, tunable hyper-parameters, or both? 🤔 Is batch normalization applied to inputs, features, weights, biases, or outputs? 🤔 How is batch normalization different from layer normalization? So I drew and calculated one entirely by hand. Goal: normalize a mini-batch of 4 examples to mean 0 and variance 1, then let the network scale it back. = 1. Given = A mini-batch of 4 training examples, each with 3 features. = 2. Linear layer = Let us multiply by the weights and add the biases. Batch norm sits after this, which answers the second question: what gets normalized is features, not inputs, weights or biases. = 3. ReLU = We apply the activation, and -2 becomes 0. Negative values are suppressed before any statistic is taken. = 4. Batch statistics = Let us compute the sum, mean, variance and standard deviation, one row at a time. A row is a feature and the four columns are the four examples, so every number here measures one feature against the rest of the batch. That is the "batch" in batch normalization, and it is exactly what layer normalization does not do. The statistics are rounded to whole numbers, which is what keeps the rest of the page doable in pen. = 5. Shift to mean 0 = We subtract the mean, in green. The four values in each feature now average to zero. = 6. Scale to variance 1 = Let us divide by the standard deviation, in orange. Each feature now has variance one, whatever scale it arrived at. = 7. Scale and shift = We multiply by a linear transformation and pass the result on. The diagonal and the last column are trainable, so having just forced every feature to mean 0 and variance 1, we hand the network the means to undo it. The outputs: Mean of each feature = [2, 1, 2] Std dev of each feature = [1, 1, 2] To the next layer = [2, -2, 2, 0], [-3, 3, 6, -3], [2, 0, 1, 2] The answers: 🤔 Both. The scale and shift are trainable, the statistics are not. Epsilon and the momentum on the running statistics are the hyper-parameters, and one mini-batch by hand needs neither. 🤔 Features, after the linear layer, not inputs, weights or biases. 🤔 Batch norm measures across the batch, one feature at a time. Layer norm measures across the features, one example at a time. 💾 Save this post!
0:17

Batch Normalization by hand ✍️ ~ 7 steps walkthrough below Batch normalization is common practice for improving training and achieving faster convergence. It sounds simple. But it is often misunderstood. 🤔 Does batch normalization involve trainable parameters, tunable hyper-parameters, or both? 🤔 Is batch normalization applied to inputs, features, weights, biases, or outputs? 🤔 How is batch normalization different from layer normalization? So I drew and calculated one entirely by hand. Goal: normalize a mini-batch of 4 examples to mean 0 and variance 1, then let the network scale it back. = 1. Given = A mini-batch of 4 training examples, each with 3 features. = 2. Linear layer = Let us multiply by the weights and add the biases. Batch norm sits after this, which answers the second question: what gets normalized is features, not inputs, weights or biases. = 3. ReLU = We apply the activation, and -2 becomes 0. Negative values are suppressed before any statistic is taken. = 4. Batch statistics = Let us compute the sum, mean, variance and standard deviation, one row at a time. A row is a feature and the four columns are the four examples, so every number here measures one feature against the rest of the batch. That is the "batch" in batch normalization, and it is exactly what layer normalization does not do. The statistics are rounded to whole numbers, which is what keeps the rest of the page doable in pen. = 5. Shift to mean 0 = We subtract the mean, in green. The four values in each feature now average to zero. = 6. Scale to variance 1 = Let us divide by the standard deviation, in orange. Each feature now has variance one, whatever scale it arrived at. = 7. Scale and shift = We multiply by a linear transformation and pass the result on. The diagonal and the last column are trainable, so having just forced every feature to mean 0 and variance 1, we hand the network the means to undo it. The outputs: Mean of each feature = [2, 1, 2] Std dev of each feature = [1, 1, 2] To the next layer = [2, -2, 2, 0], [-3, 3, 6, -3], [2, 0, 1, 2] The answers: 🤔 Both. The scale and shift are trainable, the statistics are not. Epsilon and the momentum on the running statistics are the hyper-parameters, and one mini-batch by hand needs neither. 🤔 Features, after the linear layer, not inputs, weights or biases. 🤔 Batch norm measures across the batch, one feature at a time. Layer norm measures across the features, one example at a time. 💾 Save this post!

Tom Yeh

20,848 次观看 • 2 个月前

[Dropout] by hand ✍️ Dropout is a simple yet effective way of reducing overfitting and improving generalization. This by-hand exercise lets students practice calculating dropout, thereby gaining insight into its inner workings. As an additional bonus, students get to practice calculating the gradients of the Mean Square Error (MSE) loss. After the practice, students are often surprised by how simple it is. -- 𝗡𝗲𝘁𝘄𝗼𝗿𝗸 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 -- 1. Linear(2,4) 2. ReLU 3. Dropout(0.5) 4. Linear(4,3) 5. ReLU 6. Dropout(0.33) 7. Linear(3,2) -- 𝗪𝗮𝗹𝗸𝘁𝗵𝗿𝗼𝘂𝗴𝗵 -- 🏋️ Training [1] Given ↳ A training set of 2 examples X1, X2 [2] 🟧 Random: p > 0.5 ↳ Draw 4 random numbers ↳ For each random number, if it is above 0.5, we keep and denote it as ◯. Otherwise we drop and denote it as ╳. ↳ The result is [◯, ╳, ◯, ╳] [3] 🟧 Dropout: Matrix ↳ Calculate the scaling factor: 1 / (1-p) = 2 ↳ Set the diagonal based on [◯, ╳, ◯, ╳], where ◯ = 2 and ╳ = 0 ↳ The purpose is to drop the 2nd and the 4th nodes, and scale the remaining two nodes by 2. [4] 🟦 Random: p > 0.33 ↳ Draw 3 random numbers ↳ For each random number, if it is above 0.33, we keep and denote it as ◯. Otherwise we drop and denote it as ╳. ↳ The result is [◯, ◯, ╳] [5] 🟦 Dropout: Matrix ↳ Calculate the scaling factor: 1 / (1-p) = 1.5 ↳ Set the diagonal based on [◯, ◯, ╳], where ◯ = 1.5 and ╳ = 0 ↳ The purpose is to drop the 3rd node, and scale the remaining two nodes by 1.5. [6] Feed Forward ↳ Now we have all the matrices ready across the layers, perform the feed forward pass by calculating a series of matrix multiplications from the top to the bottom ↳ ReLU activation function is applied along the way to set negative feature values to zeroes, denoted by ╳. ↳ The outputs are Y. [7] 🟥 Loss Gradients of Mean Square Error (MSE) ↳ The formula is 2 * (Y - Y') ↳ First we calculate Outputs (Y) - Targets (Y') ↳ Second we multiply each element by 2 [8] Update Weights ↳ Use loss gradients to start back propagation ↳ Update some weights (light red) ↳ The values of the new weights are for demonstration purpose only, not based on real calculation. 🔍 Inference [9] Deactivate Dropout ↳ We set both dropout matrices to identity matrices ↳ The effect is to keep all the features as is. [10] Feed Forward ↳ Take the forward pass to make predictions about unseen data.
0:10

[Dropout] by hand ✍️ Dropout is a simple yet effective way of reducing overfitting and improving generalization. This by-hand exercise lets students practice calculating dropout, thereby gaining insight into its inner workings. As an additional bonus, students get to practice calculating the gradients of the Mean Square Error (MSE) loss. After the practice, students are often surprised by how simple it is. -- 𝗡𝗲𝘁𝘄𝗼𝗿𝗸 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 -- 1. Linear(2,4) 2. ReLU 3. Dropout(0.5) 4. Linear(4,3) 5. ReLU 6. Dropout(0.33) 7. Linear(3,2) -- 𝗪𝗮𝗹𝗸𝘁𝗵𝗿𝗼𝘂𝗴𝗵 -- 🏋️ Training [1] Given ↳ A training set of 2 examples X1, X2 [2] 🟧 Random: p > 0.5 ↳ Draw 4 random numbers ↳ For each random number, if it is above 0.5, we keep and denote it as ◯. Otherwise we drop and denote it as ╳. ↳ The result is [◯, ╳, ◯, ╳] [3] 🟧 Dropout: Matrix ↳ Calculate the scaling factor: 1 / (1-p) = 2 ↳ Set the diagonal based on [◯, ╳, ◯, ╳], where ◯ = 2 and ╳ = 0 ↳ The purpose is to drop the 2nd and the 4th nodes, and scale the remaining two nodes by 2. [4] 🟦 Random: p > 0.33 ↳ Draw 3 random numbers ↳ For each random number, if it is above 0.33, we keep and denote it as ◯. Otherwise we drop and denote it as ╳. ↳ The result is [◯, ◯, ╳] [5] 🟦 Dropout: Matrix ↳ Calculate the scaling factor: 1 / (1-p) = 1.5 ↳ Set the diagonal based on [◯, ◯, ╳], where ◯ = 1.5 and ╳ = 0 ↳ The purpose is to drop the 3rd node, and scale the remaining two nodes by 1.5. [6] Feed Forward ↳ Now we have all the matrices ready across the layers, perform the feed forward pass by calculating a series of matrix multiplications from the top to the bottom ↳ ReLU activation function is applied along the way to set negative feature values to zeroes, denoted by ╳. ↳ The outputs are Y. [7] 🟥 Loss Gradients of Mean Square Error (MSE) ↳ The formula is 2 * (Y - Y') ↳ First we calculate Outputs (Y) - Targets (Y') ↳ Second we multiply each element by 2 [8] Update Weights ↳ Use loss gradients to start back propagation ↳ Update some weights (light red) ↳ The values of the new weights are for demonstration purpose only, not based on real calculation. 🔍 Inference [9] Deactivate Dropout ↳ We set both dropout matrices to identity matrices ↳ The effect is to keep all the features as is. [10] Feed Forward ↳ Take the forward pass to make predictions about unseen data.

Tom Yeh

36,034 次观看 • 2 年前

In my class, I teach the autoencoder by asking everyone to stand up. 🙆 Stretch your arms out wide. Imagine you are holding a heavy textbook (like Introduction to Algorithms by Prof. Cormen), the whole thing, every page. Now bring your hands slowly together until they almost touch your neck. The bottle "neck." The final exam is tomorrow and you are allowed one cheat sheet: whatever you can scribble on your palm. All nine hundred pages have to survive the squeeze. That is the encoder. Now imagine you sit in the exam. Push your arms back out to where they started. You try to rebuild the textbook from your palm notes. That is the decoder. Of course you cannot get every page back. What you get back is what mattered enough to write down, and the gap between the two is the loss the network is trying to shrink. I call it AI by Arms 🙆. It gets a laugh, and then it gets remembered. Goal: squeeze four numbers down to two, then rebuild the original four from them. 1. Given Let us start with four training examples: X1, X2, X3, X4. 2. Auto (copy to targets) We copy the training examples straight into the targets. That is the whole trick behind the name: "auto" is Greek for "self", and the data is its own label. 3. Encoder, layer 1 Let us multiply the inputs by the weights, add the biases, and apply ReLU. Negative values get crossed out and become zero. 4. Encoder, layer 2 (the bottleneck) We do it again, and now the four dimensions have become two. This layer is called the bottleneck, because everything has to fit through it. 5. Decoder, layer 1 Let us go back the other way: multiply, add, ReLU. This time there are no negatives to cross out. 6. Decoder, layer 2 We multiply once more and get the outputs Y. This is the decoder's attempt to rebuild the four original numbers from the two it was given. 7. MSE loss gradients Let us compare Y with the targets Y'. The gradient is 2 x (Y - Y'): subtract, then double. Those gradients kick off backpropagation, and the weights start to learn. Your entire education is all about encoding and decoding!
0:15

In my class, I teach the autoencoder by asking everyone to stand up. 🙆 Stretch your arms out wide. Imagine you are holding a heavy textbook (like Introduction to Algorithms by Prof. Cormen), the whole thing, every page. Now bring your hands slowly together until they almost touch your neck. The bottle "neck." The final exam is tomorrow and you are allowed one cheat sheet: whatever you can scribble on your palm. All nine hundred pages have to survive the squeeze. That is the encoder. Now imagine you sit in the exam. Push your arms back out to where they started. You try to rebuild the textbook from your palm notes. That is the decoder. Of course you cannot get every page back. What you get back is what mattered enough to write down, and the gap between the two is the loss the network is trying to shrink. I call it AI by Arms 🙆. It gets a laugh, and then it gets remembered. Goal: squeeze four numbers down to two, then rebuild the original four from them. 1. Given Let us start with four training examples: X1, X2, X3, X4. 2. Auto (copy to targets) We copy the training examples straight into the targets. That is the whole trick behind the name: "auto" is Greek for "self", and the data is its own label. 3. Encoder, layer 1 Let us multiply the inputs by the weights, add the biases, and apply ReLU. Negative values get crossed out and become zero. 4. Encoder, layer 2 (the bottleneck) We do it again, and now the four dimensions have become two. This layer is called the bottleneck, because everything has to fit through it. 5. Decoder, layer 1 Let us go back the other way: multiply, add, ReLU. This time there are no negatives to cross out. 6. Decoder, layer 2 We multiply once more and get the outputs Y. This is the decoder's attempt to rebuild the four original numbers from the two it was given. 7. MSE loss gradients Let us compare Y with the targets Y'. The gradient is 2 x (Y - Y'): subtract, then double. Those gradients kick off backpropagation, and the weights start to learn. Your entire education is all about encoding and decoding!

Tom Yeh

22,749 次观看 • 2 个月前

[Graph Convolutional Network] by hand ✍️ Graph Convolutional Networks (GCNs), introduced by Thomas Kipf and Max Welling in 2017, have emerged as a powerful tool in the analysis and interpretation of data structured as graphs. This exercise demonstrates how GCN works in a simple application: binary classification. -- Goal -- Predict if a node in a graph is X. -- Architecture -- 🟪 Graph Convolutional Network (GCN) 1. GCN1(4,3) 2. GCN2(3,3) 🟦 Fully Connected Network (FCN) 1. Linear1(3,5) 2. ReLU 3. Linear2(5,1) 4. Sigmoid Simplications: • Adjacent matrices are not normalized. • ReLU is applied to messages directly. -- Walkthrough -- [1] Given ↳ A graph with five nodes A, B, C, D, E [2] 🟩 Adjacency Matrix: Neighbors ↳ Add 1 for each edge to neighbors ↳ Repeat in both directions (e.g., A->C, C->A) ↳ Repeat for both GCN layers [3] 🟩 Adjacency Matrix: Self ↳ Add 1's for each self loop ↳ Equivalent to adding the identity matrix ↳ Repeat for both GCN layers [4] 🟪 GCN1: Messages ↳ Multiply the node embeddings 🟨 with weights and biases ↳ Apply ReLU (negatives → 0) ↳ The result is one message per node [5] 🟪 GCN1: Pooling ↳ Multiply the messages with the adjacent matrix ↳ The purpose is the pool messages from each node's neighbors as well as from the node itself. ↳ The result is a new feature per node [6] 🟪 GCN1: Visualize ↳ For node 1, visualize how messages are pooled to obtain a new feature for better understanding ↳ [3,0,1] + [1,0,0] = [4,0,1] [7] 🟪 GCN2: Messages ↳ Multiply the node features with weights and biases ↳ Apply ReLU (negatives → 0) ↳ The result is one message per node [8] 🟪 GCN2: Pooling ↳ Multiply the messages with the adjacent matrix ↳ The result is a new feature per node [9] 🟪 GCN2: Visualize ↳ For node 3, visualize how messages are pooled to obtain a new feature for better understanding ↳ [1,2,4] + [1,3,5] + [0,0,1] = [2,5,10] [10] 🟦 FCN: Linear 1 + ReLU ↳ Multiply node features with weights and biases ↳ Apply ReLU (negatives → 0) ↳ The result is a new feature per node ↳ Unlike in GCN layers, no messages from other nodes are included. [11] 🟦 FCN: Linear 2 ↳ Multiply node features with weights and biases [12] 🟦 FCN: Sigmoid ↳ Apply the Sigmoid activation function ↳ The purpose is to obtain a probability value for each node ↳ One way to calculate Sigmoid by hand ✍️ is to use the approximation below: • >= 3 → 1 • 0 → 0.5 • <= -3 → 0 -- Outputs -- A: 0 (Very unlikely) B: 1 (Very likely) C: 1 (Very likely) D: 1 (Very likely) E: 0.5 (Neutral)
0:12

[Graph Convolutional Network] by hand ✍️ Graph Convolutional Networks (GCNs), introduced by Thomas Kipf and Max Welling in 2017, have emerged as a powerful tool in the analysis and interpretation of data structured as graphs. This exercise demonstrates how GCN works in a simple application: binary classification. -- Goal -- Predict if a node in a graph is X. -- Architecture -- 🟪 Graph Convolutional Network (GCN) 1. GCN1(4,3) 2. GCN2(3,3) 🟦 Fully Connected Network (FCN) 1. Linear1(3,5) 2. ReLU 3. Linear2(5,1) 4. Sigmoid Simplications: • Adjacent matrices are not normalized. • ReLU is applied to messages directly. -- Walkthrough -- [1] Given ↳ A graph with five nodes A, B, C, D, E [2] 🟩 Adjacency Matrix: Neighbors ↳ Add 1 for each edge to neighbors ↳ Repeat in both directions (e.g., A->C, C->A) ↳ Repeat for both GCN layers [3] 🟩 Adjacency Matrix: Self ↳ Add 1's for each self loop ↳ Equivalent to adding the identity matrix ↳ Repeat for both GCN layers [4] 🟪 GCN1: Messages ↳ Multiply the node embeddings 🟨 with weights and biases ↳ Apply ReLU (negatives → 0) ↳ The result is one message per node [5] 🟪 GCN1: Pooling ↳ Multiply the messages with the adjacent matrix ↳ The purpose is the pool messages from each node's neighbors as well as from the node itself. ↳ The result is a new feature per node [6] 🟪 GCN1: Visualize ↳ For node 1, visualize how messages are pooled to obtain a new feature for better understanding ↳ [3,0,1] + [1,0,0] = [4,0,1] [7] 🟪 GCN2: Messages ↳ Multiply the node features with weights and biases ↳ Apply ReLU (negatives → 0) ↳ The result is one message per node [8] 🟪 GCN2: Pooling ↳ Multiply the messages with the adjacent matrix ↳ The result is a new feature per node [9] 🟪 GCN2: Visualize ↳ For node 3, visualize how messages are pooled to obtain a new feature for better understanding ↳ [1,2,4] + [1,3,5] + [0,0,1] = [2,5,10] [10] 🟦 FCN: Linear 1 + ReLU ↳ Multiply node features with weights and biases ↳ Apply ReLU (negatives → 0) ↳ The result is a new feature per node ↳ Unlike in GCN layers, no messages from other nodes are included. [11] 🟦 FCN: Linear 2 ↳ Multiply node features with weights and biases [12] 🟦 FCN: Sigmoid ↳ Apply the Sigmoid activation function ↳ The purpose is to obtain a probability value for each node ↳ One way to calculate Sigmoid by hand ✍️ is to use the approximation below: • >= 3 → 1 • 0 → 0.5 • <= -3 → 0 -- Outputs -- A: 0 (Very unlikely) B: 1 (Very likely) C: 1 (Very likely) D: 1 (Very likely) E: 0.5 (Neutral)

Tom Yeh

46,779 次观看 • 2 年前

[Deep RNN] by Hand ✍️ A Deep Recurrent Neural Network (RNN) extends a basic single-layer RNN into multiple layers of hidden states, effectively incorporating deep learning into the RNN architecture. How does a Deep RNN work? [1] Given ↳ A sequence of four inputs X1, X2, X3, X4 ⬛️ ↳ Recurrent weights and biases for hidden layers a 🟩, b 🟧, c 🟪, and the output layer y 🟦. [2] Initialize Hidden States ↳ Set a0, b0, c0 to zeros — Process X1 (t = 1)— [3] First Hidden Layer (a) 🟩: a0 → a1 ↳ The transformation matrix is horizontal concatenation of input weights, hidden state weights and biases, visualized as [⬛️ | 🟩 | ⬜️] . ↳ The state matrix is vertical concatenation of input X1, previous hidden state a0, and an extra 1, visualized as [⬛️ ; 🟩 ; 1]. ↳ Multiply the two matrices to obtain new hidden state a1 = [0 ; 1]. [4] Second Hidden Layer (b) 🟪: b0 → b1 ↳ First layer a1 🟩 becomes the input. ↳ The transformation matrix is visualized as [🟩 | 🟪 | ⬜️]. ↳ The state matrix is the combination of a1, b0, and 1, visualized as [🟩; 🟪 ; 1]. ↳ Multiply the two matrices to obtain new hidden state b1 = [1; -1]. [5] Third Hidden Layer (c) 🟧: c0 → c1 ↳ Second layer b 🟪 becomes the input. ↳ The transformation matrix is visualized as [🟪 | 🟧 | ⬜️]. ↳ The state matrix is the combination of a1, b0, and 1, visualized as [🟪; 🟧; 1]. ↳ Multiply the two matrices to obtain new hidden state b1 = [1; -1]. [6] Output Layer (Y) 🟦 ↳ The transformation matrix is visualized as [🟧 | ⬜️]. ↳ The state matrix is the combination of c0 and , visualized as [🟧; 1]. ↳ Multiply the two matrices to obtain output Y1 = [3; 0; 3]. — Process X2 (t = 2)— [7] Previous Hidden States ↳ Copy the values of a1, b1, c1. [8] Hidden 🟩🟪🟧 + Output 🟦 ↳ Repeat [3]-[6] to obtain output Y2 = [5; 0; 4] — Process X3 (t = 3)— [9] Previous Hidden States ↳ Copy the values of a2, b2, c2. [10] Hidden 🟩🟪🟧 + Output 🟦 ↳ Repeat [3]-[6] to obtain output Y3 = [13; -1; 9] — Process X4 (t = 4)— [11] Previous Hidden States ↳ Copy the values of a3, b3, c3. [12] Hidden 🟩🟪🟧 + Output 🟦 ↳ Repeat [3]-[6] to obtain output Y4 = [15; 7; 2]
0:10

[Deep RNN] by Hand ✍️ A Deep Recurrent Neural Network (RNN) extends a basic single-layer RNN into multiple layers of hidden states, effectively incorporating deep learning into the RNN architecture. How does a Deep RNN work? [1] Given ↳ A sequence of four inputs X1, X2, X3, X4 ⬛️ ↳ Recurrent weights and biases for hidden layers a 🟩, b 🟧, c 🟪, and the output layer y 🟦. [2] Initialize Hidden States ↳ Set a0, b0, c0 to zeros — Process X1 (t = 1)— [3] First Hidden Layer (a) 🟩: a0 → a1 ↳ The transformation matrix is horizontal concatenation of input weights, hidden state weights and biases, visualized as [⬛️ | 🟩 | ⬜️] . ↳ The state matrix is vertical concatenation of input X1, previous hidden state a0, and an extra 1, visualized as [⬛️ ; 🟩 ; 1]. ↳ Multiply the two matrices to obtain new hidden state a1 = [0 ; 1]. [4] Second Hidden Layer (b) 🟪: b0 → b1 ↳ First layer a1 🟩 becomes the input. ↳ The transformation matrix is visualized as [🟩 | 🟪 | ⬜️]. ↳ The state matrix is the combination of a1, b0, and 1, visualized as [🟩; 🟪 ; 1]. ↳ Multiply the two matrices to obtain new hidden state b1 = [1; -1]. [5] Third Hidden Layer (c) 🟧: c0 → c1 ↳ Second layer b 🟪 becomes the input. ↳ The transformation matrix is visualized as [🟪 | 🟧 | ⬜️]. ↳ The state matrix is the combination of a1, b0, and 1, visualized as [🟪; 🟧; 1]. ↳ Multiply the two matrices to obtain new hidden state b1 = [1; -1]. [6] Output Layer (Y) 🟦 ↳ The transformation matrix is visualized as [🟧 | ⬜️]. ↳ The state matrix is the combination of c0 and , visualized as [🟧; 1]. ↳ Multiply the two matrices to obtain output Y1 = [3; 0; 3]. — Process X2 (t = 2)— [7] Previous Hidden States ↳ Copy the values of a1, b1, c1. [8] Hidden 🟩🟪🟧 + Output 🟦 ↳ Repeat [3]-[6] to obtain output Y2 = [5; 0; 4] — Process X3 (t = 3)— [9] Previous Hidden States ↳ Copy the values of a2, b2, c2. [10] Hidden 🟩🟪🟧 + Output 🟦 ↳ Repeat [3]-[6] to obtain output Y3 = [13; -1; 9] — Process X4 (t = 4)— [11] Previous Hidden States ↳ Copy the values of a3, b3, c3. [12] Hidden 🟩🟪🟧 + Output 🟦 ↳ Repeat [3]-[6] to obtain output Y4 = [15; 7; 2]

Tom Yeh

26,562 次观看 • 2 年前

[LSTM] by Hand ✍️ LSTMs have been the most effective architecture to process long sequences of data, until our world was taken over by the Transformers. LSTMs belong to the broader family of recurrent neural network (RNNs) that process data sequentially in a recurrent manner. Transformers, on the other hand, abandon recurrence and use self-attention instead to process data concurrently in parallel. Recently, there is renewed interest in recurrence as people realized self-attention doesn’t scale to extremely long sequences, like hundreds of thousands of tokens. Mamba is a good example to bring back recurrence. All of a sudden, it is cool to study LSTMs. How do LSTMs work? [1] Given ↳ 🟨 Input sequence X1, X2, X3 (d = 3) ↳ 🟩 Hidden state h (d = 2) ↳ 🟦 Memory C (d = 2) ↳ Weight matrices Wf, Wc, Wi, Wo Process t = 1 [2] Initialize ↳ Randomly set the previous hidden state h0 to [1, 1] and memory cells C0 to [0.3, -0.5] [3] Linear Transform ↳ Multiply the four weight matrices with the concatenation of current input (X1) and the previous hidden state (h0). ↳ The results are feature values, each is a linear combination of the current input and hidden state. [4] Non-linear Transform ↳ Apply sigmoid σ to obtain gate values (between 0 and 1). • Forget gate (f1): [-4, -6] → [0, 0] • Input gate (i1): [6, 4] → [1, 1] • Output gate (o1): [4, -5] → [1, 0] ↳ Apply tanh to obtain candidate memory values (between -1 and 1) • Candidate memory (C’1): [1, -6] → [0.8, -1] [5] Update Memory ↳ Forget (C0 .* f1): Element-wise multiply the current memory with forget gate values. ↳ Input (C’1 .* o1): Element-wise multiply the “candidate” memory with input gate values. ↳ Update the memory to C1 by adding the two terms above: C0 .* f1 + C’1 .* o1 = C1 [6] Candiate Output ↳ Apply tanh to the new memory C1 to obtain candidate output o’1. [0.8, -1] → [0.7, -0.8] [7] Update Hidden State ↳ Output (o’1 .* o1 → h1): Element-wise multiply the candidate output with the output gate. ↳ The result is updated hidden state h1 ↳ Also, it is the first output. Process t = 2 [8] Initialize ↳ Copy previous hidden state h1 and memory C1 [9] Linear Transform ↳ Repeat [3] [10] Update Memory (C2) ↳ Repeat [4] and [5] [11] Update Hidden State (h2) ↳ Repeat [6] and [7] Process t = 3 [12] Initialize ↳ Copy previous hidden state h2 and memory C2 [13] Linear Transform ↳ Repeat [3] [14] Update Memory (C3) ↳ Repeat [4] and [5] [15] Update Hidden State (h3) ↳ Repeat [6] and [7]
0:15

[LSTM] by Hand ✍️ LSTMs have been the most effective architecture to process long sequences of data, until our world was taken over by the Transformers. LSTMs belong to the broader family of recurrent neural network (RNNs) that process data sequentially in a recurrent manner. Transformers, on the other hand, abandon recurrence and use self-attention instead to process data concurrently in parallel. Recently, there is renewed interest in recurrence as people realized self-attention doesn’t scale to extremely long sequences, like hundreds of thousands of tokens. Mamba is a good example to bring back recurrence. All of a sudden, it is cool to study LSTMs. How do LSTMs work? [1] Given ↳ 🟨 Input sequence X1, X2, X3 (d = 3) ↳ 🟩 Hidden state h (d = 2) ↳ 🟦 Memory C (d = 2) ↳ Weight matrices Wf, Wc, Wi, Wo Process t = 1 [2] Initialize ↳ Randomly set the previous hidden state h0 to [1, 1] and memory cells C0 to [0.3, -0.5] [3] Linear Transform ↳ Multiply the four weight matrices with the concatenation of current input (X1) and the previous hidden state (h0). ↳ The results are feature values, each is a linear combination of the current input and hidden state. [4] Non-linear Transform ↳ Apply sigmoid σ to obtain gate values (between 0 and 1). • Forget gate (f1): [-4, -6] → [0, 0] • Input gate (i1): [6, 4] → [1, 1] • Output gate (o1): [4, -5] → [1, 0] ↳ Apply tanh to obtain candidate memory values (between -1 and 1) • Candidate memory (C’1): [1, -6] → [0.8, -1] [5] Update Memory ↳ Forget (C0 .* f1): Element-wise multiply the current memory with forget gate values. ↳ Input (C’1 .* o1): Element-wise multiply the “candidate” memory with input gate values. ↳ Update the memory to C1 by adding the two terms above: C0 .* f1 + C’1 .* o1 = C1 [6] Candiate Output ↳ Apply tanh to the new memory C1 to obtain candidate output o’1. [0.8, -1] → [0.7, -0.8] [7] Update Hidden State ↳ Output (o’1 .* o1 → h1): Element-wise multiply the candidate output with the output gate. ↳ The result is updated hidden state h1 ↳ Also, it is the first output. Process t = 2 [8] Initialize ↳ Copy previous hidden state h1 and memory C1 [9] Linear Transform ↳ Repeat [3] [10] Update Memory (C2) ↳ Repeat [4] and [5] [11] Update Hidden State (h2) ↳ Repeat [6] and [7] Process t = 3 [12] Initialize ↳ Copy previous hidden state h2 and memory C2 [13] Linear Transform ↳ Repeat [3] [14] Update Memory (C3) ↳ Repeat [4] and [5] [15] Update Hidden State (h3) ↳ Repeat [6] and [7]

Tom Yeh

73,030 次观看 • 2 年前

MLP in PyTorch by hand ✍️ ~ 7 steps walkthrough below Goal: fill in every blank in the PyTorch code to build a multi-layer perceptron. 1. Given Let us start with a code template on the left and the network it is supposed to build on the right. Every blank in the code can be worked out from the picture. 2. Linear layer We count: 3 features in, 4 features out. So the weight matrix is 4 by 3. There is an extra column for the biases, which means bias = T. 3. ReLU Let us apply the activation. ReLU crosses out the negatives, so -1 becomes 0. 4. Linear layer The input size is 4, because that is what the previous layer put out. The output size is 2. A 2 by 4 weight matrix, and this time no extra column, so bias = F. 5. ReLU We cross out the negatives again. 6. Linear layer Two features in, five out. A 5 by 2 weight matrix, with a bias column, so bias = T. 7. Sigmoid Let us finish. Sigmoid squashes the raw scores (3, 0, -2, 5, -5) into probabilities between 0 and 1. You have just implemented a three-layer deep neural network by hand. ✍️ == Story == Three years ago I gave this exercise to my students, to connect the code to the math. They found it odd. Every other AI course they were taking lived inside a Jupyter notebook, and here I was handing out paper. Three years later, my colleagues are the ones rushing to move their materials to paper. The exercise has not changed. Paper still asks the one thing a notebook lets you skip: do you actually understand what the code is doing? If you can tell me why the weight matrix is 4 by 3, and why bias is F on the second layer, you understand nn.Linear better than someone who has been copy-pasting it for a year. 💾 Save this post! #AIbyHand #PyTorch #DeepLearning
0:14

MLP in PyTorch by hand ✍️ ~ 7 steps walkthrough below Goal: fill in every blank in the PyTorch code to build a multi-layer perceptron. 1. Given Let us start with a code template on the left and the network it is supposed to build on the right. Every blank in the code can be worked out from the picture. 2. Linear layer We count: 3 features in, 4 features out. So the weight matrix is 4 by 3. There is an extra column for the biases, which means bias = T. 3. ReLU Let us apply the activation. ReLU crosses out the negatives, so -1 becomes 0. 4. Linear layer The input size is 4, because that is what the previous layer put out. The output size is 2. A 2 by 4 weight matrix, and this time no extra column, so bias = F. 5. ReLU We cross out the negatives again. 6. Linear layer Two features in, five out. A 5 by 2 weight matrix, with a bias column, so bias = T. 7. Sigmoid Let us finish. Sigmoid squashes the raw scores (3, 0, -2, 5, -5) into probabilities between 0 and 1. You have just implemented a three-layer deep neural network by hand. ✍️ == Story == Three years ago I gave this exercise to my students, to connect the code to the math. They found it odd. Every other AI course they were taking lived inside a Jupyter notebook, and here I was handing out paper. Three years later, my colleagues are the ones rushing to move their materials to paper. The exercise has not changed. Paper still asks the one thing a notebook lets you skip: do you actually understand what the code is doing? If you can tell me why the weight matrix is 4 by 3, and why bias is F on the second layer, you understand nn.Linear better than someone who has been copy-pasting it for a year. 💾 Save this post! #AIbyHand #PyTorch #DeepLearning

Tom Yeh

13,318 次观看 • 3 个月前

Transformer by hand ✍️ ~ 6 steps walkthrough below Open the hood of a transformer and the parts list is overwhelming: embeddings, positional encoding, attention weighting, self-attention, cross-attention, multi-head attention, layer norm, skip connections, softmax, linear, Nx, shifted right, query, key, value, masking. Which of those actually make the car run? Two of them. Attention weighting and the feed-forward network. Everything else is an enhancement to make it run faster and longer, which is how we got from a car to a truck, and to the word "large" in large language model. So I drew and calculated those two parts entirely by hand. Goal: push five features through one transformer block, filling in every cell yourself. 1. Given Five positions of input features, arriving from the previous block. 2. Attention matrix Let us feed all five features to a query-key module (QK) and read back an attention weight matrix, A. The details of that module are a post of their own. 3. Attention weighting We multiply the input features by A to get the attention weighted features, Z. Still five positions. The effect is to combine features *across positions*, horizontally: X1 becomes X1 + X2, X2 becomes X2 + X3, and so on. 4. First layer Let us feed all five weighted features into the first layer of the FFN. Multiply by the weights and biases. This time the combining happens *across feature dimensions*, vertically, and each feature grows from 3 numbers to 4. Note that every position goes through the same weight matrix. That is what "position-wise" means. 5. ReLU We cross out the negatives. They become zeros. 6. Second layer Let us bring it back down: 4 dimensions to 3. The output feeds the next block, which has a completely separate set of parameters, and the whole thing runs again. You have just calculated a transformer block by hand. ✍️ The takeaway: the two parts are doing two different jobs, and neither one alone is enough. Attention mixes *across positions*, so a feature can see its neighbours. The FFN mixes *across feature dimensions*, so each position can think about itself. Horizontal, then vertical. Then that pattern repeats N times, each block with its own separate set of weights. That is the Nx from the list up top, and that is what makes the transformer run. 💾 Save this post! #AIbyHand #Transformers #DeepLearning
0:12

Transformer by hand ✍️ ~ 6 steps walkthrough below Open the hood of a transformer and the parts list is overwhelming: embeddings, positional encoding, attention weighting, self-attention, cross-attention, multi-head attention, layer norm, skip connections, softmax, linear, Nx, shifted right, query, key, value, masking. Which of those actually make the car run? Two of them. Attention weighting and the feed-forward network. Everything else is an enhancement to make it run faster and longer, which is how we got from a car to a truck, and to the word "large" in large language model. So I drew and calculated those two parts entirely by hand. Goal: push five features through one transformer block, filling in every cell yourself. 1. Given Five positions of input features, arriving from the previous block. 2. Attention matrix Let us feed all five features to a query-key module (QK) and read back an attention weight matrix, A. The details of that module are a post of their own. 3. Attention weighting We multiply the input features by A to get the attention weighted features, Z. Still five positions. The effect is to combine features *across positions*, horizontally: X1 becomes X1 + X2, X2 becomes X2 + X3, and so on. 4. First layer Let us feed all five weighted features into the first layer of the FFN. Multiply by the weights and biases. This time the combining happens *across feature dimensions*, vertically, and each feature grows from 3 numbers to 4. Note that every position goes through the same weight matrix. That is what "position-wise" means. 5. ReLU We cross out the negatives. They become zeros. 6. Second layer Let us bring it back down: 4 dimensions to 3. The output feeds the next block, which has a completely separate set of parameters, and the whole thing runs again. You have just calculated a transformer block by hand. ✍️ The takeaway: the two parts are doing two different jobs, and neither one alone is enough. Attention mixes *across positions*, so a feature can see its neighbours. The FFN mixes *across feature dimensions*, so each position can think about itself. Horizontal, then vertical. Then that pattern repeats N times, each block with its own separate set of weights. That is the Nx from the list up top, and that is what makes the transformer run. 💾 Save this post! #AIbyHand #Transformers #DeepLearning

Tom Yeh

26,211 次观看 • 3 个月前

[SVM] by Hand ✍️ Support Vector Machines (SVMs) reigned supreme in machine learning before the ascendancy of the deep learning revolution. This exercise compares Linear vs RBF SVMs---how they classify two test vectors, after SVMs are trained from six training vectors. [1] Given xi: Six training vectors (blue rows 🟦) yi: Labels Using xi and yi, we learned ai and b (red borders): ↳ai: coefficient for each training vector i. • Non-zero: A Support Vector that defines the decision boundary • Zero: Too far from the decision boundary, ignored ↳b: bias (how much the decision boundary should be shifted) x’j: Two test vectors (yellow columns 🟨) (To simplify hand calculation, training and test vectors are not normalized.) ~ 𝗟𝗶𝗻𝗲𝗮𝗿 𝗦𝗩𝗠 ~ 🟩 Kernel Matrix (K) [2]-[3] [2] Test Vector 1 ↳ Take dot product between the test vector 🟨 and every training vector 🟦 ↳ The dot product approximates the “cosine similarity” between two vectors ↳ Output: 1st column of K [3] Test Vector 2 ↳ Similar to [2] ↳ Output: 2nd column of K 🟧 Decision Boundary [4]-[6] [4] Unsigned Coefficients → Signed Weights ↳ Multiply each coefficient with the corresponding label ↳ The 2nd training vector is NOT a support vector because its coefficient is 0. [5] Weighted Combination ↳ Multiply weights and bias with K ↳ Output: “signed” distance to the decision boundary ... X'1: (2)*13+0+(1)*12+(-1)*15+(1)*19+(-1)*27+(2) = 17 X'2: (2)*2+0+(1)*3+(-1)*4+(1)*8+(-1)*8+(2) = 5 ... [6] Classify ↳ Take the sign ... X'1: 1 > 0 → Positive + X'2: 5 > 0 → Positive + ... ~ 𝗥𝗕𝗙 𝗦𝗩𝗠 ~ Given ai: Learned coefficients b: Learned bias 🟩 Kernel Matrix (K) [7]-[15] 𝘛𝘦𝘴𝘵 𝘝𝘦𝘤𝘵𝘰𝘳 (𝘟’1) 🟨 L2 Distance 🟪 [7]-[9] [7] Squared Difference ... i=1: (1-2)^2=1, (2-4)^2=4, (1-3)^2=4 ... [8] Sum [9] Square Root [10] Negative Scaling ↳ Multiply by -1: Note that L2 is a distance metric. The negation converts distance to similarity. ↳ Multiply by gamma γ: The purpose is to control how much influence each training example has. A small gamma means each training example pulls the decision boundary more lightly, resulting in smoother decision boundaries. ↳ The result is “negative scaled L2” [11] Exponentiate ↳ Raise e to the power of the “negative scaled L2” ↳ Use the provided table to look up the value of e^ ↳ Output: The 1st column of K 𝘛𝘦𝘴𝘵 𝘝𝘦𝘤𝘵𝘰𝘳 2 (𝘟’2) 🟨 L2 Distance 🟪 [12]-[14] [12] Squared Difference [13] Sum [14] Square Root [15] Negative Scaling [16] Exponentiate ↳Output: The 2nd column of K 🟧 Decision Boundary [17]-[19] [17] Unsigned Coefficients → Signed Weights [18] Weighted Combination ... X'1: (20)*0.7+(-10)*0.8+(10)*0.9+(-10)*0.7+0+(-10)*0.9+(-1) = -2 X'2: (20)*0.9+(-10)*0.7+(10)*0.8+(-10)*0.8+0+(-10)*0.9+(-1) = 1 ... [19] Classify ... X'1: -2 0 → + ... ↳Output: The 2nd column of K 🟧 Decision Boundary [17]-[19] [17] Unsigned Coefficients → Signed Weights [18] Weighted Combination ... X'1: (20)*0.7+(-10)*0.8+(10)*0.9+(-10)*0.7+0+(-10)*0.9+(-1) = -2 X'2: (20)*0.9+(-10)*0.7+(10)*0.8+(-10)*0.8+0+(-10)*0.9+(-1) = 1 ... [19] Classify ... X'1: -2 0 → + ...
0:19

[SVM] by Hand ✍️ Support Vector Machines (SVMs) reigned supreme in machine learning before the ascendancy of the deep learning revolution. This exercise compares Linear vs RBF SVMs---how they classify two test vectors, after SVMs are trained from six training vectors. [1] Given xi: Six training vectors (blue rows 🟦) yi: Labels Using xi and yi, we learned ai and b (red borders): ↳ai: coefficient for each training vector i. • Non-zero: A Support Vector that defines the decision boundary • Zero: Too far from the decision boundary, ignored ↳b: bias (how much the decision boundary should be shifted) x’j: Two test vectors (yellow columns 🟨) (To simplify hand calculation, training and test vectors are not normalized.) ~ 𝗟𝗶𝗻𝗲𝗮𝗿 𝗦𝗩𝗠 ~ 🟩 Kernel Matrix (K) [2]-[3] [2] Test Vector 1 ↳ Take dot product between the test vector 🟨 and every training vector 🟦 ↳ The dot product approximates the “cosine similarity” between two vectors ↳ Output: 1st column of K [3] Test Vector 2 ↳ Similar to [2] ↳ Output: 2nd column of K 🟧 Decision Boundary [4]-[6] [4] Unsigned Coefficients → Signed Weights ↳ Multiply each coefficient with the corresponding label ↳ The 2nd training vector is NOT a support vector because its coefficient is 0. [5] Weighted Combination ↳ Multiply weights and bias with K ↳ Output: “signed” distance to the decision boundary ... X'1: (2)*13+0+(1)*12+(-1)*15+(1)*19+(-1)*27+(2) = 17 X'2: (2)*2+0+(1)*3+(-1)*4+(1)*8+(-1)*8+(2) = 5 ... [6] Classify ↳ Take the sign ... X'1: 1 > 0 → Positive + X'2: 5 > 0 → Positive + ... ~ 𝗥𝗕𝗙 𝗦𝗩𝗠 ~ Given ai: Learned coefficients b: Learned bias 🟩 Kernel Matrix (K) [7]-[15] 𝘛𝘦𝘴𝘵 𝘝𝘦𝘤𝘵𝘰𝘳 (𝘟’1) 🟨 L2 Distance 🟪 [7]-[9] [7] Squared Difference ... i=1: (1-2)^2=1, (2-4)^2=4, (1-3)^2=4 ... [8] Sum [9] Square Root [10] Negative Scaling ↳ Multiply by -1: Note that L2 is a distance metric. The negation converts distance to similarity. ↳ Multiply by gamma γ: The purpose is to control how much influence each training example has. A small gamma means each training example pulls the decision boundary more lightly, resulting in smoother decision boundaries. ↳ The result is “negative scaled L2” [11] Exponentiate ↳ Raise e to the power of the “negative scaled L2” ↳ Use the provided table to look up the value of e^ ↳ Output: The 1st column of K 𝘛𝘦𝘴𝘵 𝘝𝘦𝘤𝘵𝘰𝘳 2 (𝘟’2) 🟨 L2 Distance 🟪 [12]-[14] [12] Squared Difference [13] Sum [14] Square Root [15] Negative Scaling [16] Exponentiate ↳Output: The 2nd column of K 🟧 Decision Boundary [17]-[19] [17] Unsigned Coefficients → Signed Weights [18] Weighted Combination ... X'1: (20)*0.7+(-10)*0.8+(10)*0.9+(-10)*0.7+0+(-10)*0.9+(-1) = -2 X'2: (20)*0.9+(-10)*0.7+(10)*0.8+(-10)*0.8+0+(-10)*0.9+(-1) = 1 ... [19] Classify ... X'1: -2 0 → + ... ↳Output: The 2nd column of K 🟧 Decision Boundary [17]-[19] [17] Unsigned Coefficients → Signed Weights [18] Weighted Combination ... X'1: (20)*0.7+(-10)*0.8+(10)*0.9+(-10)*0.7+0+(-10)*0.9+(-1) = -2 X'2: (20)*0.9+(-10)*0.7+(10)*0.8+(-10)*0.8+0+(-10)*0.9+(-1) = 1 ... [19] Classify ... X'1: -2 0 → + ...

Tom Yeh

35,708 次观看 • 2 年前

[Discrete Fourier Transform] by Hand ✍️ In signal processing, the Discrete Fourier Transform (DFT) is no doubt the most important method. But the math involved is extremely complex, literally, involving a summation over a complex number term e^(-iwt). I developed this exercise to demonstrate that underneath such complexity, DFT is just a series of matrix multiplications you can calculate by hand. ✍️ Once you see that, it should not surprise you that a deep neural network, which is also a series of matrix multiplications, with activation functions in-between, can learn to perform DFT to process and analyze signals so effectively. How does DFT work? [1] Given ↳ Signals A, B, and C in the 🟧 frequency domain: ◦ A = cos(w) + 2cos(2w) ◦ B = cos(w) + cos(3w) + cos(4w) ◦ C = -cos(2w) + cos(3w) ◦ Each signal is a weighed sum of four cosine waves at frequencies 1w, 2w, 3w, and 4w. ◦ We will apply Inverse DFT to convert the signals to time domain representations, and then demonstrate DFT can convert back to their original frequency domain representations. ↳ Signal X in the 🟩 time domain. X is sampled at 10 time points 1t, 2t, …, 10t: ◦ X = [-2.5, -1.8, 3, -0.7, -1.0, -0.7, 3, -1.8, -2.5, 5] ◦ Suppose X is also a weighted sum of the same four cosine waves, but we don’t already know their weights. We will apply DFT to discover them. [2] 🟧 Frequency Matrix (F) ↳ Write the coefficients of A, B, C as a matrix F. Each signal is a row. Each frequency is a column. ↳ A → [1, 2, 0, 0] ↳ B → [1, 0, 1, 1] ↳ C → [0, 1-, 1, 0] [3] Cosine → Discrete ↳ Sample from the continuous cosine waves at discrete time points 1t, 2t, 3t, to 10t. [4] Cosine Matrix (W) ↳ Write the samples as a matrix, Each frequency is a row. Each time point is a column. [5] Inverse DFT: 🟧 Frequency → 🟩 Time ↳ Multiply the frequency matrix F and the cosine matrix W. ↳ The meaning of this multiplication is to linearly combine the four cosine waves (rows in W) into time-domain signals (rows in T) using the weights specified in F. ↳ The result is matrix T, which are signals A, B, C converted to the time domain. Each signal is a row. Each time point is a column. [6] Transpose ↳ Transpose T, converting each signal’s time domain representation from a row to a column. [7] DFT: 🟩 Time → 🟧 Frequency ↳ Multiply the cosine matrix W with the transpose of matrix T. ↳ The purpose of this multiplication is to take a dot-product between each time-domain signal (columns in the transpose of T) and each cosine wave (rows in W), which has the effect of projecting the signal onto a cosine wave to determine how much they are correlated. Zero means not correlated at all. ↳ The result is an intermediate version of the “recovered” frequency matrix where each column corresponds to a signal and each row corresponds to a frequency. ↳ Compared to the original frequency matrix F, this intermediate matrix has non-zero weights in the correct places, but scaled up by a factor of 5 (n/2, n=10). For example, signal A, originally [1,2,0,0], is recovered at [5,10,0,0]. [8] Scale ↳ Multiply each value by 2/n = 1/5 to scale down the intermediate matrix to match the magnitude of the original frequency matrix F. [9] Transpose ↳ Transpose the recovered frequency matrix back to the same orientation of the original frequency matrix F. ↳ Like magic 🪄, the result is identical to the original F, which means DFT successfully recovered the frequency components of signals A, B, C. [10] Apply DFT to X: 🟩 Time → 🟧 Frequency ↳ Now that we have some confidence in DFT’s ability to recover frequency components, we apply DFT to X’s time-domain representation by multiplying W with X. ↳ The result is the an intermediate matrix. [11] Scale ↳ Similarly, we scale down by a factor of 5 to obtain the recovered frequency components of X (a column). [12] Transpose ↳ Similarly, we transpose the recovered column to row to match the orientation of the frequency matrix. ↳ Using the coefficients [0,0,3,2], we can write the equation of X as 3cos(3w) + 2cos(4w). Notes: I hope this by hand exercise helps you understand the essence of DFT. But there is more technical details, such as: • Sine: The complete DFT math also includes sine waves that follow a similar calculation process. • Phase: Here, we assume all the cosine waves are aligned at the origin, namely, phase is 0. If a phase p is added, for example, cos(w+p), we will need to calculate the sine component and use their ratio to figure out what p is. • Magnitude: If phase is not zero, the magnitude will need to be calculated by combining both cosine and sine terms.
0:12

[Discrete Fourier Transform] by Hand ✍️ In signal processing, the Discrete Fourier Transform (DFT) is no doubt the most important method. But the math involved is extremely complex, literally, involving a summation over a complex number term e^(-iwt). I developed this exercise to demonstrate that underneath such complexity, DFT is just a series of matrix multiplications you can calculate by hand. ✍️ Once you see that, it should not surprise you that a deep neural network, which is also a series of matrix multiplications, with activation functions in-between, can learn to perform DFT to process and analyze signals so effectively. How does DFT work? [1] Given ↳ Signals A, B, and C in the 🟧 frequency domain: ◦ A = cos(w) + 2cos(2w) ◦ B = cos(w) + cos(3w) + cos(4w) ◦ C = -cos(2w) + cos(3w) ◦ Each signal is a weighed sum of four cosine waves at frequencies 1w, 2w, 3w, and 4w. ◦ We will apply Inverse DFT to convert the signals to time domain representations, and then demonstrate DFT can convert back to their original frequency domain representations. ↳ Signal X in the 🟩 time domain. X is sampled at 10 time points 1t, 2t, …, 10t: ◦ X = [-2.5, -1.8, 3, -0.7, -1.0, -0.7, 3, -1.8, -2.5, 5] ◦ Suppose X is also a weighted sum of the same four cosine waves, but we don’t already know their weights. We will apply DFT to discover them. [2] 🟧 Frequency Matrix (F) ↳ Write the coefficients of A, B, C as a matrix F. Each signal is a row. Each frequency is a column. ↳ A → [1, 2, 0, 0] ↳ B → [1, 0, 1, 1] ↳ C → [0, 1-, 1, 0] [3] Cosine → Discrete ↳ Sample from the continuous cosine waves at discrete time points 1t, 2t, 3t, to 10t. [4] Cosine Matrix (W) ↳ Write the samples as a matrix, Each frequency is a row. Each time point is a column. [5] Inverse DFT: 🟧 Frequency → 🟩 Time ↳ Multiply the frequency matrix F and the cosine matrix W. ↳ The meaning of this multiplication is to linearly combine the four cosine waves (rows in W) into time-domain signals (rows in T) using the weights specified in F. ↳ The result is matrix T, which are signals A, B, C converted to the time domain. Each signal is a row. Each time point is a column. [6] Transpose ↳ Transpose T, converting each signal’s time domain representation from a row to a column. [7] DFT: 🟩 Time → 🟧 Frequency ↳ Multiply the cosine matrix W with the transpose of matrix T. ↳ The purpose of this multiplication is to take a dot-product between each time-domain signal (columns in the transpose of T) and each cosine wave (rows in W), which has the effect of projecting the signal onto a cosine wave to determine how much they are correlated. Zero means not correlated at all. ↳ The result is an intermediate version of the “recovered” frequency matrix where each column corresponds to a signal and each row corresponds to a frequency. ↳ Compared to the original frequency matrix F, this intermediate matrix has non-zero weights in the correct places, but scaled up by a factor of 5 (n/2, n=10). For example, signal A, originally [1,2,0,0], is recovered at [5,10,0,0]. [8] Scale ↳ Multiply each value by 2/n = 1/5 to scale down the intermediate matrix to match the magnitude of the original frequency matrix F. [9] Transpose ↳ Transpose the recovered frequency matrix back to the same orientation of the original frequency matrix F. ↳ Like magic 🪄, the result is identical to the original F, which means DFT successfully recovered the frequency components of signals A, B, C. [10] Apply DFT to X: 🟩 Time → 🟧 Frequency ↳ Now that we have some confidence in DFT’s ability to recover frequency components, we apply DFT to X’s time-domain representation by multiplying W with X. ↳ The result is the an intermediate matrix. [11] Scale ↳ Similarly, we scale down by a factor of 5 to obtain the recovered frequency components of X (a column). [12] Transpose ↳ Similarly, we transpose the recovered column to row to match the orientation of the frequency matrix. ↳ Using the coefficients [0,0,3,2], we can write the equation of X as 3cos(3w) + 2cos(4w). Notes: I hope this by hand exercise helps you understand the essence of DFT. But there is more technical details, such as: • Sine: The complete DFT math also includes sine waves that follow a similar calculation process. • Phase: Here, we assume all the cosine waves are aligned at the origin, namely, phase is 0. If a phase p is added, for example, cos(w+p), we will need to calculate the sine component and use their ratio to figure out what p is. • Magnitude: If phase is not zero, the magnitude will need to be calculated by combining both cosine and sine terms.

Tom Yeh

116,622 次观看 • 2 年前

Vector Database by Hand ✍️ Vector databases are revolutionizing how we search and analyze complex data. They have become the backbone of Retrieval Augmented Generation (#RAG). How do vector databases work? [1] Given ↳ A dataset of three sentences, each has 3 words (or tokens) ↳ In practice, a dataset may contain millions or billions of sentences. The max number of tokens may be tens of thousands (e.g., 32,768 mistral-7b). Process "how are you" [2] 🟨 Word Embeddings ↳ For each word, look up corresponding word embedding vector from a table of 22 vectors, where 22 is the vocabulary size. ↳ In practice, the vocabulary size can be tens of thousands. The word embedding dimensions are in the thousands (e.g., 1024, 4096) [3] 🟩 Encoding ↳ Feed the sequence of word embeddings to an encoder to obtain a sequence of feature vectors, one per word. ↳ Here, the encoder is a simple one layer perceptron (linear layer + ReLU) ↳ In practice, the encoder is a transformer or one of its many variants. [4] 🟩 Mean Pooling ↳ Merge the sequence of feature vectors into a single vector using "mean pooling" which is to average across the columns. ↳ The result is a single vector. We often call it "text embeddings" or "sentence embeddings." ↳ Other pooling techniques are possible, such as CLS. But mean pooling is the most common. [5] 🟦 Indexing ↳ Reduce the dimensions of the text embedding vector by a projection matrix. The reduction rate is 50% (4->2). ↳ In practice, the values in this projection matrix is much more random. ↳ The purpose is similar to that of hashing, which is to obtain a short representation to allow faster comparison and retrieval. ↳ The resulting dimension-reduced index vector is saved in the vector storage. [6] Process "who are you" ↳ Repeat [2]-[5] [7] Process "who am I" ↳ Repeat [2]-[5] Now we have indexed our dataset in the vector database. [8] 🟥 Query: "am I you" ↳ Repeat [2]-[5] ↳ The result is a 2-d query vector. [9] 🟥 Dot Products ↳ Take dot product between the query vector and database vectors. They are all 2-d. ↳ The purpose is to use dot product to estimate similarity. ↳ By transposing the query vector, this step becomes a matrix multiplication. [10] 🟥 Nearest Neighbor ↳ Find the largest dot product by linear scan. ↳ The sentence with the highest dot product is "who am I" ↳ In practice, because scanning billions of vectors is slow, we use an Approximate Nearest Neighbor (ANN) algorithm like the Hierarchical Navigable Small Worlds (HNSW).
0:17

Vector Database by Hand ✍️ Vector databases are revolutionizing how we search and analyze complex data. They have become the backbone of Retrieval Augmented Generation (#RAG). How do vector databases work? [1] Given ↳ A dataset of three sentences, each has 3 words (or tokens) ↳ In practice, a dataset may contain millions or billions of sentences. The max number of tokens may be tens of thousands (e.g., 32,768 mistral-7b). Process "how are you" [2] 🟨 Word Embeddings ↳ For each word, look up corresponding word embedding vector from a table of 22 vectors, where 22 is the vocabulary size. ↳ In practice, the vocabulary size can be tens of thousands. The word embedding dimensions are in the thousands (e.g., 1024, 4096) [3] 🟩 Encoding ↳ Feed the sequence of word embeddings to an encoder to obtain a sequence of feature vectors, one per word. ↳ Here, the encoder is a simple one layer perceptron (linear layer + ReLU) ↳ In practice, the encoder is a transformer or one of its many variants. [4] 🟩 Mean Pooling ↳ Merge the sequence of feature vectors into a single vector using "mean pooling" which is to average across the columns. ↳ The result is a single vector. We often call it "text embeddings" or "sentence embeddings." ↳ Other pooling techniques are possible, such as CLS. But mean pooling is the most common. [5] 🟦 Indexing ↳ Reduce the dimensions of the text embedding vector by a projection matrix. The reduction rate is 50% (4->2). ↳ In practice, the values in this projection matrix is much more random. ↳ The purpose is similar to that of hashing, which is to obtain a short representation to allow faster comparison and retrieval. ↳ The resulting dimension-reduced index vector is saved in the vector storage. [6] Process "who are you" ↳ Repeat [2]-[5] [7] Process "who am I" ↳ Repeat [2]-[5] Now we have indexed our dataset in the vector database. [8] 🟥 Query: "am I you" ↳ Repeat [2]-[5] ↳ The result is a 2-d query vector. [9] 🟥 Dot Products ↳ Take dot product between the query vector and database vectors. They are all 2-d. ↳ The purpose is to use dot product to estimate similarity. ↳ By transposing the query vector, this step becomes a matrix multiplication. [10] 🟥 Nearest Neighbor ↳ Find the largest dot product by linear scan. ↳ The sentence with the highest dot product is "who am I" ↳ In practice, because scanning billions of vectors is slow, we use an Approximate Nearest Neighbor (ANN) algorithm like the Hierarchical Navigable Small Worlds (HNSW).

Tom Yeh

192,216 次观看 • 2 年前

ReLU vs Leaky ReLU 👉 = ReLU = ReLU is the default activation in modern deep learning — cheap to compute, and stable enough to train networks hundreds of layers deep. To see what it does, picture five boba tea shops on the same block — 𝚊, 𝚋, 𝚌, 𝚍, 𝚎 — each running their own books. Each value is a shop's monthly profit — receipts minus rent, ingredients, and wages. When profit is positive, the shop stays open and the owner pockets every dollar. When profit turns negative, the shop runs out of cash and shutters — the lights go off, the books are wiped to zero. ReLU is exactly that rule, applied one shop at a time. Read the diagram left to right. The first column is the raw value x — each shop's profit at month's end. The second column is the gate: 1 if the shop is open (x > 0), 0 if it has shuttered. The last column is the ReLU output: open shops pass their profit through untouched, while shuttered ones are zeroed out. Five rows means five parallel shops on the same block, each evaluated independently. That's why ReLU is called an element-wise activation: every neuron decides its own fate. = LeakyRelu = Plain ReLU wipes negative values to zero — clean, but a shop that shutters can never recover, since both its output and its gradient stay pinned at zero. This is the dying ReLU problem, and in deep networks it can quietly kill a meaningful fraction of the units. Leaky ReLU is the one-line fix: instead of shuttering, the shop files for Chapter 11 protection and keeps the lights on at reduced capacity. Its debt is restructured down to a fraction α (typically 0.1) — the rest is forgiven, and the shop is wounded, not killed. A small negative signal still flows through, so the gradient survives, and the shop can crawl back to life if a TikTok goes viral. Read the diagram left to right. The first column is the raw value x — each shop's profit at month's end. The second column is the leakage α — the fraction of the loss held over after restructuring (default 0.1, editable). The third column is the gate: 1 for shops still in the black, α for those operating under bankruptcy protection. The last column is the Leaky ReLU output: y = x · gate. Profitable shops pass through untouched; struggling ones shrink by a factor of α but still carry a sign. Five rows means five parallel shops, each evaluated independently. Like ReLU, this is an element-wise activation: every neuron's fate is decided on its own merits. #aibyhahd
0:18

ReLU vs Leaky ReLU 👉 = ReLU = ReLU is the default activation in modern deep learning — cheap to compute, and stable enough to train networks hundreds of layers deep. To see what it does, picture five boba tea shops on the same block — 𝚊, 𝚋, 𝚌, 𝚍, 𝚎 — each running their own books. Each value is a shop's monthly profit — receipts minus rent, ingredients, and wages. When profit is positive, the shop stays open and the owner pockets every dollar. When profit turns negative, the shop runs out of cash and shutters — the lights go off, the books are wiped to zero. ReLU is exactly that rule, applied one shop at a time. Read the diagram left to right. The first column is the raw value x — each shop's profit at month's end. The second column is the gate: 1 if the shop is open (x > 0), 0 if it has shuttered. The last column is the ReLU output: open shops pass their profit through untouched, while shuttered ones are zeroed out. Five rows means five parallel shops on the same block, each evaluated independently. That's why ReLU is called an element-wise activation: every neuron decides its own fate. = LeakyRelu = Plain ReLU wipes negative values to zero — clean, but a shop that shutters can never recover, since both its output and its gradient stay pinned at zero. This is the dying ReLU problem, and in deep networks it can quietly kill a meaningful fraction of the units. Leaky ReLU is the one-line fix: instead of shuttering, the shop files for Chapter 11 protection and keeps the lights on at reduced capacity. Its debt is restructured down to a fraction α (typically 0.1) — the rest is forgiven, and the shop is wounded, not killed. A small negative signal still flows through, so the gradient survives, and the shop can crawl back to life if a TikTok goes viral. Read the diagram left to right. The first column is the raw value x — each shop's profit at month's end. The second column is the leakage α — the fraction of the loss held over after restructuring (default 0.1, editable). The third column is the gate: 1 for shops still in the black, α for those operating under bankruptcy protection. The last column is the Leaky ReLU output: y = x · gate. Profitable shops pass through untouched; struggling ones shrink by a factor of α but still carry a sign. Five rows means five parallel shops, each evaluated independently. Like ReLU, this is an element-wise activation: every neuron's fate is decided on its own merits. #aibyhahd

Tom Yeh

32,561 次观看 • 5 个月前

Microsoft made 100B parameter models run on a single CPU. bitnet.cpp: The official inference framework for 1-bit LLMs. The math behind 1-bit LLMs is what makes them revolutionary. Traditional LLMs use 16-bit floating point weights. Every parameter is a number like 0.0023847 or -1.4729. When you run inference, you multiply these floats together. Billions of times. That's why you need GPUs, they're optimized for floating point matrix multiplication. BitNet b1.58 uses ternary weights: {-1, 0, 1}. That's not a simplification. That's a fundamental change in the math. When your weights are only -1, 0, or 1: → Multiply by 1 = keep the value → Multiply by -1 = flip the sign → Multiply by 0 = skip entirely Matrix multiplication becomes addition and subtraction. No floating point operations. No GPU required. This is why bitnet.cpp achieves: → 2.37x to 6.17x speedup on x86 CPUs → 1.37x to 5.07x speedup on ARM CPUs → 71.9% to 82.2% energy reduction on x86 → 55.4% to 70.0% energy reduction on ARM The speedups scale with model size. Larger models see bigger gains because there are more operations to simplify. A 100B parameter model running at human reading speed (5-7 tokens/second) on a single CPU. That's not optimization. That's a different paradigm. Why 1.58 bits? Because log₂(3) ≈ 1.58. Three possible values = 1.58 bits of information per weight. The key insight: These models aren't quantized after training. They're trained from scratch with ternary weights. The model learns to work within the constraint. No precision loss. No quality tradeoff.
0:13

Microsoft made 100B parameter models run on a single CPU. bitnet.cpp: The official inference framework for 1-bit LLMs. The math behind 1-bit LLMs is what makes them revolutionary. Traditional LLMs use 16-bit floating point weights. Every parameter is a number like 0.0023847 or -1.4729. When you run inference, you multiply these floats together. Billions of times. That's why you need GPUs, they're optimized for floating point matrix multiplication. BitNet b1.58 uses ternary weights: {-1, 0, 1}. That's not a simplification. That's a fundamental change in the math. When your weights are only -1, 0, or 1: → Multiply by 1 = keep the value → Multiply by -1 = flip the sign → Multiply by 0 = skip entirely Matrix multiplication becomes addition and subtraction. No floating point operations. No GPU required. This is why bitnet.cpp achieves: → 2.37x to 6.17x speedup on x86 CPUs → 1.37x to 5.07x speedup on ARM CPUs → 71.9% to 82.2% energy reduction on x86 → 55.4% to 70.0% energy reduction on ARM The speedups scale with model size. Larger models see bigger gains because there are more operations to simplify. A 100B parameter model running at human reading speed (5-7 tokens/second) on a single CPU. That's not optimization. That's a different paradigm. Why 1.58 bits? Because log₂(3) ≈ 1.58. Three possible values = 1.58 bits of information per weight. The key insight: These models aren't quantized after training. They're trained from scratch with ternary weights. The model learns to work within the constraint. No precision loss. No quality tradeoff.

Tech with Mak

23,202 次观看 • 6 个月前

1-bit Qwen3.8-27B just more than DOUBLED its MMLU-Pro score. They didn’t change the decoder. 29.04% -> 61.54%, according to a new paper from ISTA-DASLab. Same lab behind GSQ-RCO, some of the best tiny Qwen3.8-27B quants in my book. Their method is called Disaggregated Quantization. You use different weights to read the prompt and generate the answer. A trained NVFP4 prefiller processes your input and builds the cache. Then the tiny GGUF decoder takes over. They train the prefiller specifically for that decoder, so it learns to produce representations the heavily compressed model can use. The extra checkpoint is 12.8 GiB, but you don’t need all of it in VRAM. Stream it from SSD layer by layer, reusing GPU memory. With a long enough prompt, loading can overlap computation. That’s a pretty fucking good reason to pay attention if your GPU is short on memory. I want to test this on my 4x3090 rig with NVMe. The released implementation targets Blackwell, so this needs adaptation. First I’ll check whether the quality gains survive on Ampere, then measure whether offloading actually helps. Paper:
0:17

1-bit Qwen3.8-27B just more than DOUBLED its MMLU-Pro score. They didn’t change the decoder. 29.04% -> 61.54%, according to a new paper from ISTA-DASLab. Same lab behind GSQ-RCO, some of the best tiny Qwen3.8-27B quants in my book. Their method is called Disaggregated Quantization. You use different weights to read the prompt and generate the answer. A trained NVFP4 prefiller processes your input and builds the cache. Then the tiny GGUF decoder takes over. They train the prefiller specifically for that decoder, so it learns to produce representations the heavily compressed model can use. The extra checkpoint is 12.8 GiB, but you don’t need all of it in VRAM. Stream it from SSD layer by layer, reusing GPU memory. With a long enough prompt, loading can overlap computation. That’s a pretty fucking good reason to pay attention if your GPU is short on memory. I want to test this on my 4x3090 rig with NVMe. The released implementation targets Blackwell, so this needs adaptation. First I’ll check whether the quality gains survive on Ampere, then measure whether offloading actually helps. Paper:

Alexey Fateev

19,870 次观看 • 19 天前

Colmap 4.0 was very recently released, so it inspired me to do some work to better understand it and its new capabilities with Rerun. I want to really understand how Colmap, and in particular, pycolmap, works outside of just calling it via the CLI. So my goal is to use the low-level pycolmap API to log every part of the pipeline. The explicit goal is to have an alternative to the SQLite database that I can utilize. Instead of SQLite, I want to try logging everything directly to rerun and use RRD. This means I can have deep inspectability and still save the features/matches/2D view geometry, but be able to view it directly in rerun. I think this is one of the superpowers that rerun provides; data and visualizations are deeply integrated. As I'm often working with sequential data (videos), I'm going to specifically focus on four things: 1. Monocular Video Simple: Calls high-level APIs such as pycolmap.extract_features, pycolmap.match_sequential, pycolmap.incremental_mapping. These are basically identical to the CLI options and provide a good baseline. 2. Monocular Video Streamed: Take the above high-level APIs and break them down to their iterator version, logging each component in a streamed manner. This way, I can stream the intermediate features to rerun while the extraction/matching/mapping is happening. 3. Rig with unknown calibration: <- WHAT THE VIDEO SHOWS This is probably the most interesting version and the first one I've been working on. It allows one to set a rig between known sensors, such as in VR/AR devices, leading to much better reconstructions with multiple cameras. This is the case where we don't know the calibration a priori, so we have to run a reconstruction twice: once as a normal Colmap reconstruction with no rig constraints, use this to generate the constraints, and then do it again with the newly found rig. 4. Rig with known calibration: This is the RoboCap example, where we have a pre-calibrated set of sensors, so we don't need to run the two reconstructions and also gain better matching between cameras, both spatially and temporally. Again, this leads to a much better reconstruction! Along with all this, GLOMAP has become a first-class global mapper, making it super easy to use directly within pycolmap! I'm excited to do more with this and compare it to things like pycuvslam, vipe, and other alternatives.
0:20

Colmap 4.0 was very recently released, so it inspired me to do some work to better understand it and its new capabilities with Rerun. I want to really understand how Colmap, and in particular, pycolmap, works outside of just calling it via the CLI. So my goal is to use the low-level pycolmap API to log every part of the pipeline. The explicit goal is to have an alternative to the SQLite database that I can utilize. Instead of SQLite, I want to try logging everything directly to rerun and use RRD. This means I can have deep inspectability and still save the features/matches/2D view geometry, but be able to view it directly in rerun. I think this is one of the superpowers that rerun provides; data and visualizations are deeply integrated. As I'm often working with sequential data (videos), I'm going to specifically focus on four things: 1. Monocular Video Simple: Calls high-level APIs such as pycolmap.extract_features, pycolmap.match_sequential, pycolmap.incremental_mapping. These are basically identical to the CLI options and provide a good baseline. 2. Monocular Video Streamed: Take the above high-level APIs and break them down to their iterator version, logging each component in a streamed manner. This way, I can stream the intermediate features to rerun while the extraction/matching/mapping is happening. 3. Rig with unknown calibration: <- WHAT THE VIDEO SHOWS This is probably the most interesting version and the first one I've been working on. It allows one to set a rig between known sensors, such as in VR/AR devices, leading to much better reconstructions with multiple cameras. This is the case where we don't know the calibration a priori, so we have to run a reconstruction twice: once as a normal Colmap reconstruction with no rig constraints, use this to generate the constraints, and then do it again with the newly found rig. 4. Rig with known calibration: This is the RoboCap example, where we have a pre-calibrated set of sensors, so we don't need to run the two reconstructions and also gain better matching between cameras, both spatially and temporally. Again, this leads to a much better reconstruction! Along with all this, GLOMAP has become a first-class global mapper, making it super easy to use directly within pycolmap! I'm excited to do more with this and compare it to things like pycuvslam, vipe, and other alternatives.

Pablo Vela

30,070 次观看 • 6 个月前

Dear Nigerians, "Patriotism means to stand by the country. It does not mean to stand by the president." ~Theodore Roosevelt On this national scale, if the mass number of Nigerians are suffering, their lives and properties in jeopardy, it is okay to speak up, call a spade a spade and stop this gov from leading us astray. Now more than ever, we must unite. We must unite against the ruiners of Nigeria before they bury what is left of her. Organization is the first primary weapon of the oppressed. This is the only weapon we are yet to explore, and we can not explore this until we unite. The oppressed Nigerians must organize to form a force against the ruiners of Nigeria and take back our country to have a system that is for the benefit of the Nigerian people and not a few in the position of power. President will come and go, only the people will remain. Don't give those in power today to play politics with your future. Let's unite, not necessary as one but as a team. A team with different skills, languages culture, religion and tradition, but one CAUSE regardless of our tribal differences to stand against the common enemy who have now shown to be a curse on our very own existence as an amazing people of this great nation. ~OIA✊🏿
0:15

Dear Nigerians, "Patriotism means to stand by the country. It does not mean to stand by the president." ~Theodore Roosevelt On this national scale, if the mass number of Nigerians are suffering, their lives and properties in jeopardy, it is okay to speak up, call a spade a spade and stop this gov from leading us astray. Now more than ever, we must unite. We must unite against the ruiners of Nigeria before they bury what is left of her. Organization is the first primary weapon of the oppressed. This is the only weapon we are yet to explore, and we can not explore this until we unite. The oppressed Nigerians must organize to form a force against the ruiners of Nigeria and take back our country to have a system that is for the benefit of the Nigerian people and not a few in the position of power. President will come and go, only the people will remain. Don't give those in power today to play politics with your future. Let's unite, not necessary as one but as a team. A team with different skills, languages culture, religion and tradition, but one CAUSE regardless of our tribal differences to stand against the common enemy who have now shown to be a curse on our very own existence as an amazing people of this great nation. ~OIA✊🏿

𝐎i𝐀

15,608 次观看 • 1 年前

A Letter to Our Community: The Road Ahead for Robotics To our Community and Partners, As we step into 2026, our mission at Axis is clearer than ever: Constructing the definitive End-to-End Scaling Layer for Robotics. Our goal is to accelerate the transfer of diverse human intelligence into Robotics General Intelligence (RGI). By owning the critical path of intelligence creation, we are turning the physical limitations of robotics into a scalable, software-driven future. Here is our strategic outlook and roadmap for the year ahead. The Core Thesis: Simulation is the Only Way Out The path to RGI is currently blocked by Data Scarcity, Generalization Fragility, and Hardware Fragmentation. At Axis, we believe Simulation is the only way out. Our Simulation Data Platform and Data Augmentation Engine transform raw data into "Synthetic Gold". Backed by academic milestones like Roboverse, Skill Blending, and GraspVLA, we have proven that pure simulation can achieve the generalization required for the real world. We don’t just collect data; we architect it. The Engine: Why Crypto? We believe RGI should come from all, not a few. Crypto is not just a feature; it is the primitive that powers our entire ecosystem flywheel: - Incentive Mechanism: Democratizing contribution and rewarding the trainers and developers. - Assetization: Turning proprietary data and refined models into liquid, ownable assets. - Verifiable Workflow: We are opening the "Black Box" of AI. By bringing total transparency to the Task Generation → Data Collection → Model Training pipeline, we ensure every byte of intelligence is verifiable, traceable, and secure. 2026 Strategic Deliverables This year, we are committed to delivering three foundational pillars: - The World's Largest Training Dataset for Robots: A robot training set—diverse, high-quality interaction data at an unprecedented scale. - A Robotics Foundation Model: A universal robotic brain trained on our pure simulation and synthetic data, capable of robust cross-embodiment transfer and open-world adaptability. - Evolvable Robot Hardware: Robots deployed with Axis models that autonomously evolve through continuous interaction, turning every deployment into a self-improving node within our RGI network. The Ultimate Vision We are building more than models; we are architecting the Distributed Machine Economy. A future where every dataset, model, and robotic embodiment is a verifiable asset in a global, autonomous network. Thank you for building the future of intelligence with us✌️📷
0:13

A Letter to Our Community: The Road Ahead for Robotics To our Community and Partners, As we step into 2026, our mission at Axis is clearer than ever: Constructing the definitive End-to-End Scaling Layer for Robotics. Our goal is to accelerate the transfer of diverse human intelligence into Robotics General Intelligence (RGI). By owning the critical path of intelligence creation, we are turning the physical limitations of robotics into a scalable, software-driven future. Here is our strategic outlook and roadmap for the year ahead. The Core Thesis: Simulation is the Only Way Out The path to RGI is currently blocked by Data Scarcity, Generalization Fragility, and Hardware Fragmentation. At Axis, we believe Simulation is the only way out. Our Simulation Data Platform and Data Augmentation Engine transform raw data into "Synthetic Gold". Backed by academic milestones like Roboverse, Skill Blending, and GraspVLA, we have proven that pure simulation can achieve the generalization required for the real world. We don’t just collect data; we architect it. The Engine: Why Crypto? We believe RGI should come from all, not a few. Crypto is not just a feature; it is the primitive that powers our entire ecosystem flywheel: - Incentive Mechanism: Democratizing contribution and rewarding the trainers and developers. - Assetization: Turning proprietary data and refined models into liquid, ownable assets. - Verifiable Workflow: We are opening the "Black Box" of AI. By bringing total transparency to the Task Generation → Data Collection → Model Training pipeline, we ensure every byte of intelligence is verifiable, traceable, and secure. 2026 Strategic Deliverables This year, we are committed to delivering three foundational pillars: - The World's Largest Training Dataset for Robots: A robot training set—diverse, high-quality interaction data at an unprecedented scale. - A Robotics Foundation Model: A universal robotic brain trained on our pure simulation and synthetic data, capable of robust cross-embodiment transfer and open-world adaptability. - Evolvable Robot Hardware: Robots deployed with Axis models that autonomously evolve through continuous interaction, turning every deployment into a self-improving node within our RGI network. The Ultimate Vision We are building more than models; we are architecting the Distributed Machine Economy. A future where every dataset, model, and robotic embodiment is a verifiable asset in a global, autonomous network. Thank you for building the future of intelligence with us✌️📷

Axis Robotics

28,096 次观看 • 9 个月前

Break-A-Scene: Extracting Multiple Concepts from a Single Image introduce the task of textual scene decomposition: given a single image of a scene that may contain several concepts, we aim to extract a distinct text token for each concept, enabling fine-grained control over the generated scenes. To this end, we propose augmenting the input image with masks that indicate the presence of target concepts. These masks can be provided by the user or generated automatically by a pre-trained segmentation model. We then present a novel two-phase customization process that optimizes a set of dedicated textual embeddings (handles), as well as the model weights, striking a delicate balance between accurately capturing the concepts and avoiding overfitting. We employ a masked diffusion loss to enable handles to generate their assigned concepts, complemented by a novel loss on cross-attention maps to prevent entanglement. We also introduce union-sampling, a training strategy aimed to improve the ability of combining multiple concepts in generated images. We use several automatic metrics to quantitatively compare our method against several baselines, and further affirm the results using a user study. Finally, we showcase several applications of our method paper page:
0:12

Break-A-Scene: Extracting Multiple Concepts from a Single Image introduce the task of textual scene decomposition: given a single image of a scene that may contain several concepts, we aim to extract a distinct text token for each concept, enabling fine-grained control over the generated scenes. To this end, we propose augmenting the input image with masks that indicate the presence of target concepts. These masks can be provided by the user or generated automatically by a pre-trained segmentation model. We then present a novel two-phase customization process that optimizes a set of dedicated textual embeddings (handles), as well as the model weights, striking a delicate balance between accurately capturing the concepts and avoiding overfitting. We employ a masked diffusion loss to enable handles to generate their assigned concepts, complemented by a novel loss on cross-attention maps to prevent entanglement. We also introduce union-sampling, a training strategy aimed to improve the ability of combining multiple concepts in generated images. We use several automatic metrics to quantitatively compare our method against several baselines, and further affirm the results using a user study. Finally, we showcase several applications of our method paper page:

AK

154,511 次观看 • 3 年前

Here’s how the Book of Enoch tells you when judgment, or the Geophysical Event, is near: The years become shorter, the rain is withheld, fruit ripens late, and the Moon and stars depart from their established order. This section is among the oldest parts of the book, and copies of it were found in Cave 4 at Qumran. So, what does this mean that even an AI like Grok fails to understand? Let me explain it to you... The Moon and stars going “out of order” is connected to axial and apsidal precession. The rising of the stars that once announced the seasons gradually ceases to align with the same seasonal events. For example, in ancient Egypt, the heliacal rising of Sirius, its first appearance in the dawn sky after a period of invisibility, announced the approaching annual flooding of the Nile and the beginning of the Egyptian New Year. However, because of axial precession, this rising would eventually no longer announce the flooding at the same time. The celestial markers slowly shift in relation to the seasons. Because of this slow change in the Earth’s orientation toward the Sun, along with small changes in the distance between the Earth and the Sun, the climate begins to change and shift gradually. However, these changes build toward a tipping point, after which the climate completely collapses and reorganizes. That is how the Sahara became a desert around 6,000 years ago. These climatic cycles are known as Bond events. Hapgood suggested that these slow changes affect the buildup and melting of ice, which could then potentially cause the Earth’s outer crust to slip, triggering the biblical flood or causing the oceans to spill across the continents! Today’s climate change is connected to axial and apsidal precession and the approach toward a tipping point. The Book of Enoch is correct. We are close to Judgment! I made a short video to show you how the stars change their positions relative to the Moon, the Sun, and the five visible planets over a long period of time. That is why the Book of Enoch has been sidelined by the elites. They do not want you to know the truth!
0:13

Here’s how the Book of Enoch tells you when judgment, or the Geophysical Event, is near: The years become shorter, the rain is withheld, fruit ripens late, and the Moon and stars depart from their established order. This section is among the oldest parts of the book, and copies of it were found in Cave 4 at Qumran. So, what does this mean that even an AI like Grok fails to understand? Let me explain it to you... The Moon and stars going “out of order” is connected to axial and apsidal precession. The rising of the stars that once announced the seasons gradually ceases to align with the same seasonal events. For example, in ancient Egypt, the heliacal rising of Sirius, its first appearance in the dawn sky after a period of invisibility, announced the approaching annual flooding of the Nile and the beginning of the Egyptian New Year. However, because of axial precession, this rising would eventually no longer announce the flooding at the same time. The celestial markers slowly shift in relation to the seasons. Because of this slow change in the Earth’s orientation toward the Sun, along with small changes in the distance between the Earth and the Sun, the climate begins to change and shift gradually. However, these changes build toward a tipping point, after which the climate completely collapses and reorganizes. That is how the Sahara became a desert around 6,000 years ago. These climatic cycles are known as Bond events. Hapgood suggested that these slow changes affect the buildup and melting of ice, which could then potentially cause the Earth’s outer crust to slip, triggering the biblical flood or causing the oceans to spill across the continents! Today’s climate change is connected to axial and apsidal precession and the approach toward a tipping point. The Book of Enoch is correct. We are close to Judgment! I made a short video to show you how the stars change their positions relative to the Moon, the Sun, and the five visible planets over a long period of time. That is why the Book of Enoch has been sidelined by the elites. They do not want you to know the truth!

Open Minded Approach

157,300 次观看 • 2 个月前

I've been working a lot with SAM3 and the Momentum Human Rig (MHR). I finally integrated it into the data I'm working with Rerun. The progression I've taken looks as follows SAM3 + SAM3D-body on 1. a single image 2. a set of multiple images 3. a single video 4. A multiview video capture I took inspiration from the SAM3D-body paper and built a multiview fitting optimization pipeline. This pipeline involves using the 2D keypoints from the single-view pipeline, triangulating them, and employing an L1 loss between the 2D/3D keypoints. The temporal stability isn't great, so that's the next portion I'm going to focus on. One really frustrating thing about SAM3D-body is the lack of per-joint confidence values. It makes it harder to deal with occlusions. I'm probably going to need to use a separate model, or maybe add a confidence head.
0:10

I've been working a lot with SAM3 and the Momentum Human Rig (MHR). I finally integrated it into the data I'm working with Rerun. The progression I've taken looks as follows SAM3 + SAM3D-body on 1. a single image 2. a set of multiple images 3. a single video 4. A multiview video capture I took inspiration from the SAM3D-body paper and built a multiview fitting optimization pipeline. This pipeline involves using the 2D keypoints from the single-view pipeline, triangulating them, and employing an L1 loss between the 2D/3D keypoints. The temporal stability isn't great, so that's the next portion I'm going to focus on. One really frustrating thing about SAM3D-body is the lack of per-joint confidence values. It makes it harder to deal with occlusions. I'm probably going to need to use a separate model, or maybe add a confidence head.

Pablo Vela

43,436 次观看 • 9 个月前