Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Autoencoder by hand✍️Excel~ I designed this exercise to show how an Encoder-Decoder network convert input to code and reconstruct input from code. It is annotated with equations, PyTorch, and graphs. I also made a medium version.👇Join the 'AI Math' community. Download xlsx.

Tom Yeh

65,560 subscribers

54,516 views • 1 year ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

ResNet by hand✍️Excel~ I designed this exercise to compare a ResNet to an MLP and show skip-connections are simply identity matrices next to weights and biases. I also made a medium version to show how it scales. 👇Join the 'AI Math' community. Download xlsx.

ResNet by hand✍️Excel~ I designed this exercise to compare a ResNet to an MLP and show skip-connections are simply identity matrices next to weights and biases. I also made a medium version to show how it scales. 👇Join the 'AI Math' community. Download xlsx.

Tom Yeh

29,925 views • 1 year ago

Transformer by hand✍️ Excel ~ I designed this exercise to show the core math of a Transformer model is to combine columns (attention), combine rows (feed forward), and repeat.👇Join the 'AI Math' community. 👇Download xlsx.

Transformer by hand✍️ Excel ~ I designed this exercise to show the core math of a Transformer model is to combine columns (attention), combine rows (feed forward), and repeat.👇Join the 'AI Math' community. 👇Download xlsx.

Tom Yeh

66,816 views • 1 year ago

AlphaFold by hand✍️ Excel ~ I designed this exercise to show (1) MSA multi-head attention, (2) Pair triangular update, two key components of the EvoFormer architecture.👇Join the AI Math community. Download xlsx.

AlphaFold by hand✍️ Excel ~ I designed this exercise to show (1) MSA multi-head attention, (2) Pair triangular update, two key components of the EvoFormer architecture.👇Join the AI Math community. Download xlsx.

Tom Yeh

105,032 views • 1 year ago

Backpropagation by hand✍️ ~ spreadsheet. I designed this exercise to show it is possible to calculate backpropagation for a non-trivial, three-layer network by hand. p.s. I just started this community to share useful resources on AI math. 👇 Join the community.

Backpropagation by hand✍️ ~ spreadsheet. I designed this exercise to show it is possible to calculate backpropagation for a non-trivial, three-layer network by hand. p.s. I just started this community to share useful resources on AI math. 👇 Join the community.

Tom Yeh

121,911 views • 2 years ago

Transformer: Multi-Head Attention ~ Math vs Code 🔢💻 ~ I made this visualization to show you how to implement the multi-head attention math in PyTorch within 50 LoC. Multi-Head Attention is what makes the Transformer's performance outstanding. It captures and represents more diverse linguistic relationships and patterns, and attends to different learned input embedding spaces. The parallel computing design also makes the model more efficient.

Transformer: Multi-Head Attention ~ Math vs Code 🔢💻 ~ I made this visualization to show you how to implement the multi-head attention math in PyTorch within 50 LoC. Multi-Head Attention is what makes the Transformer's performance outstanding. It captures and represents more diverse linguistic relationships and patterns, and attends to different learned input embedding spaces. The parallel computing design also makes the model more efficient.

Yan Chen

33,326 views • 1 year ago

[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 data. The International Conference on Learning Representations (ICLR) this year announced its first ever "Test of Time Award" to recognizes the VAE paper, published 10 years ago. This exercise demonstrates how to calculate a VAE by hand. [1] Given: ↳ Three training examples X1, X2, X3 ↳ Copy training examples to the bottom ↳ The purpose is to train the network to reconstruct the training examples. ↳ Since each target is a training example itself, we use the Greek word "auto" which means "self." This crucial step is what makes an autoencoder "auto." [2] Encoder: Layer 1 + ReLU ↳ Multiply inputs with weights and biases ↳ Apply ReLU, crossing out negative values (-1 -> 0) [3] Encoder: Mean and Variance ↳ Multiply features with two sets of weights and biases ↳ 🟩 The first set predicts the means (𝜇) of latent distributions ↳ 🟪 The second set predicts the standard deviation (𝜎) of latent distributions [4] Reparameterization Trick: Random Offset ↳ Sample epsilon ε from the normal distribution with mean = 0 and variance = 1. ↳ The purpose is to randomly pick a offset away from the mean. ↳ Multiply the standard deviation values with epsilon values. ↳ The purpose is to scale the offset by the standard deviation. [5] Reparameterization Trick: Mean + Offset ↳ Add the sampled offset to predicted mean ↳ The result are new parameters or features 🟨 as inputs to the Decoder. [6] Decoder: Layer 1 + ReLU ↳ Multiply input features with weights and biases ↳ Apply ReLU, crossing out negative values. Here, -4 is crossed out. [7] Decoder: Layer 2 ↳ Multiply features with weights and biases ↳ The output is Decoder's attempt to reconstruct the input data X from reparameterized distributions described by 𝜇 and 𝜎. [8]-[10] KL Divergence Loss [8] Loss Gradient: Mean 𝜇 ↳ We want 𝜇 to approach 0. ↳ A lot of math called SGVB simplifies the calculation of loss gradients to simply 𝜇 [9,10] Loss Gradient: Stdev 𝜎 ↳ We want 𝜎 to approach 1. ↳ A lot of math simplifies the calculation to 𝜎 - (1/ 𝜎) [11] Reconstruction Loss ↳ We want the reconstructed data Y (dark 🟧) to be the same as the input data X. ↳ Some math involving Mean Square Error simplifies the calculation to Y - X.

[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 data. The International Conference on Learning Representations (ICLR) this year announced its first ever "Test of Time Award" to recognizes the VAE paper, published 10 years ago. This exercise demonstrates how to calculate a VAE by hand. [1] Given: ↳ Three training examples X1, X2, X3 ↳ Copy training examples to the bottom ↳ The purpose is to train the network to reconstruct the training examples. ↳ Since each target is a training example itself, we use the Greek word "auto" which means "self." This crucial step is what makes an autoencoder "auto." [2] Encoder: Layer 1 + ReLU ↳ Multiply inputs with weights and biases ↳ Apply ReLU, crossing out negative values (-1 -> 0) [3] Encoder: Mean and Variance ↳ Multiply features with two sets of weights and biases ↳ 🟩 The first set predicts the means (𝜇) of latent distributions ↳ 🟪 The second set predicts the standard deviation (𝜎) of latent distributions [4] Reparameterization Trick: Random Offset ↳ Sample epsilon ε from the normal distribution with mean = 0 and variance = 1. ↳ The purpose is to randomly pick a offset away from the mean. ↳ Multiply the standard deviation values with epsilon values. ↳ The purpose is to scale the offset by the standard deviation. [5] Reparameterization Trick: Mean + Offset ↳ Add the sampled offset to predicted mean ↳ The result are new parameters or features 🟨 as inputs to the Decoder. [6] Decoder: Layer 1 + ReLU ↳ Multiply input features with weights and biases ↳ Apply ReLU, crossing out negative values. Here, -4 is crossed out. [7] Decoder: Layer 2 ↳ Multiply features with weights and biases ↳ The output is Decoder's attempt to reconstruct the input data X from reparameterized distributions described by 𝜇 and 𝜎. [8]-[10] KL Divergence Loss [8] Loss Gradient: Mean 𝜇 ↳ We want 𝜇 to approach 0. ↳ A lot of math called SGVB simplifies the calculation of loss gradients to simply 𝜇 [9,10] Loss Gradient: Stdev 𝜎 ↳ We want 𝜎 to approach 1. ↳ A lot of math simplifies the calculation to 𝜎 - (1/ 𝜎) [11] Reconstruction Loss ↳ We want the reconstructed data Y (dark 🟧) to be the same as the input data X. ↳ Some math involving Mean Square Error simplifies the calculation to Y - X.

Tom Yeh

48,475 views • 2 years ago

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

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 views • 2 months ago

When Ghostty detects a password input prompt, it now changes the cursor to a lock and on macOS enables the secure input API. When the secure input API is enabled, we show a neat, animated icon that explains what's going on when clicked. Another example of native UI wins (imo). Secure Input is the macOS system API that prevents accessibility APIs from reading your keystrokes, so things like screen recording software and so on can't read your passwords. Other terminals on macOS support secure input. I think only iTerm also supports secure input on password detection. So as a disclaimer, I'm not trying to claim this as a huge innovation, I just think our implementation is nice. 😊

When Ghostty detects a password input prompt, it now changes the cursor to a lock and on macOS enables the secure input API. When the secure input API is enabled, we show a neat, animated icon that explains what's going on when clicked. Another example of native UI wins (imo). Secure Input is the macOS system API that prevents accessibility APIs from reading your keystrokes, so things like screen recording software and so on can't read your passwords. Other terminals on macOS support secure input. I think only iTerm also supports secure input on password detection. So as a disclaimer, I'm not trying to claim this as a huge innovation, I just think our implementation is nice. 😊

Mitchell Hashimoto

81,772 views • 2 years ago

At MIT, I learned about RNNs in my NLP class with Prof. Michael Collins. He built a model from my keystrokes to predict who I was. To me, it felt like a magic box. Years later, when I had to teach RNNs, I forced myself to go inside the box. ⬇️ Download: First, with a tiny example on paper by hand ✍️. Then, a slightly larger one in Excel. That’s when it finally clicked: 👉 The weights are reused (weight matrices on the left side) 👉 The hidden states are passed down (H's) When I built it by hand and saw everything visually, it clicked, just math you can actually trace. Now I try to give others the same “aha!” moment I had. ⬇️ Download Excel:

At MIT, I learned about RNNs in my NLP class with Prof. Michael Collins. He built a model from my keystrokes to predict who I was. To me, it felt like a magic box. Years later, when I had to teach RNNs, I forced myself to go inside the box. ⬇️ Download: First, with a tiny example on paper by hand ✍️. Then, a slightly larger one in Excel. That’s when it finally clicked: 👉 The weights are reused (weight matrices on the left side) 👉 The hidden states are passed down (H's) When I built it by hand and saw everything visually, it clicked, just math you can actually trace. Now I try to give others the same “aha!” moment I had. ⬇️ Download Excel:

Tom Yeh

220,876 views • 11 months ago

I FOUND AN INPUT FOR RAGING DEMON (I cant believe I'm actually labbing in MH...) When there's a wound > ⏹️+⏹️ +🔼 +⏺️ Here's also my camera focusing on the controller to show how it works THIS IS CRAZY, ARE WE PLAYING SF6?!

I FOUND AN INPUT FOR RAGING DEMON (I cant believe I'm actually labbing in MH...) When there's a wound > ⏹️+⏹️ +🔼 +⏺️ Here's also my camera focusing on the controller to show how it works THIS IS CRAZY, ARE WE PLAYING SF6?!

Shincry

468,017 views • 1 year ago

I wrote down my thoughts on how AI turned me from a designer to a builder, how I’ve kept my taste, and how the process naturally pushes me to understand code I also have this totally unrelated ascii editor at the top if you like having something to fidget with while reading

I wrote down my thoughts on how AI turned me from a designer to a builder, how I’ve kept my taste, and how the process naturally pushes me to understand code I also have this totally unrelated ascii editor at the top if you like having something to fidget with while reading

Fay

50,905 views • 4 months ago

ONE PROMPT. AN ENTIRE FINANCIAL DASHBOARD. I asked a new AI agent to build it, and it created this interactive dashboard from scratch. It planned the project, wrote the code, ran it locally, and handled the final touches. Meet EvoX Agent 🧵👇

ONE PROMPT. AN ENTIRE FINANCIAL DASHBOARD. I asked a new AI agent to build it, and it created this interactive dashboard from scratch. It planned the project, wrote the code, ran it locally, and handled the final touches. Meet EvoX Agent 🧵👇

gus

60,661 views • 1 day ago

I designed these worksheets to turn Agentic AI concepts into simple math problems you can do by hand. ✍️ Download PDF: Problems 1 to 5: 1. Count the tokens: split on spaces, one word per box 2. Subword splitting: when one word is three tokens 3. Punctuation counts: the marks are tokens too 4. Tokens per word: the ratio that turns words into a bill 5. Will it fit? A document against a context window Why am I making these worksheets? AI is making people (including me) think less. It's just too easy to ask AI a few questions about a new AI concept and start to believe I get it. Until recently, we could use a coding problem to practice a new AI concept. But now it has become too easy to ask AI to write the code, and we start to think we must know the concept, since we technically solved a coding problem. Thus, my approach is to recast AI concepts as simple math problems we must solve with pen and paper. Using our hands is one way to motivate ourselves to start thinking again. ✍️ ~ Prof. Tom Yeh

I designed these worksheets to turn Agentic AI concepts into simple math problems you can do by hand. ✍️ Download PDF: Problems 1 to 5: 1. Count the tokens: split on spaces, one word per box 2. Subword splitting: when one word is three tokens 3. Punctuation counts: the marks are tokens too 4. Tokens per word: the ratio that turns words into a bill 5. Will it fit? A document against a context window Why am I making these worksheets? AI is making people (including me) think less. It's just too easy to ask AI a few questions about a new AI concept and start to believe I get it. Until recently, we could use a coding problem to practice a new AI concept. But now it has become too easy to ask AI to write the code, and we start to think we must know the concept, since we technically solved a coding problem. Thus, my approach is to recast AI concepts as simple math problems we must solve with pen and paper. Using our hands is one way to motivate ourselves to start thinking again. ✍️ ~ Prof. Tom Yeh

Tom Yeh

18,291 views • 1 month ago

We (Claude and I) made a new package: voxmlx It's an MLX implementation of Mistral's Voxtral mini realtime speech recognition model. It supports streaming audio and runs pretty fast on a laptop. To use it, simply: `uvx voxmlx` Also I did not write a single line of code for this package. Every line of code was written by Claude Code. More on that in thread.

We (Claude and I) made a new package: voxmlx It's an MLX implementation of Mistral's Voxtral mini realtime speech recognition model. It supports streaming audio and runs pretty fast on a laptop. To use it, simply: `uvx voxmlx` Also I did not write a single line of code for this package. Every line of code was written by Claude Code. More on that in thread.

Awni Hannun

33,896 views • 7 months ago

Here’s half a dozen telesnaps fed to a AI with no other input. Yes, it’s obviously an uncanny mess, and I know AI stuff is a controversial topic, but this route for recons definitely has future potential as image-to-video continues to improve.

Here’s half a dozen telesnaps fed to a AI with no other input. Yes, it’s obviously an uncanny mess, and I know AI stuff is a controversial topic, but this route for recons definitely has future potential as image-to-video continues to improve.

Gav Rymill 🤖

103,617 views • 3 years ago

The future of AI in web development? This app generates code and images to convert a screenshot into HTML/Tailwind CSS. I just gave it a screenshot of my X feed, and look what it produced (Notes & link in the comments):

The future of AI in web development? This app generates code and images to convert a screenshot into HTML/Tailwind CSS. I just gave it a screenshot of my X feed, and look what it produced (Notes & link in the comments):

Alvaro Cintas

303,534 views • 2 years ago

Here is the tool I built yesterday in Claude code in maybe 10-20 minutes? It truly baffles me when people say “the AI bubble” this simple tool would’ve took devs 2-5 days and would have cost me 500$-1000$ Claude code made it for me on a 20$ monthly subscription, and a free Gemini api key. I actually want to preform an experiment asking developers how much they would charge for such a tool to get a consensus on how much money Claude Code has saved me.

Here is the tool I built yesterday in Claude code in maybe 10-20 minutes? It truly baffles me when people say “the AI bubble” this simple tool would’ve took devs 2-5 days and would have cost me 500$-1000$ Claude code made it for me on a 20$ monthly subscription, and a free Gemini api key. I actually want to preform an experiment asking developers how much they would charge for such a tool to get a consensus on how much money Claude Code has saved me.

Chris

170,273 views • 8 months ago

i found the code for this in the files!! it's in an object called obj_bushjump, it's a bit broken by default due to the assets being moved at some point and a bit of the code being fucked with/deleted for some reason, but i was able to fix it

i found the code for this in the files!! it's in an object called obj_bushjump, it's a bit broken by default due to the assets being moved at some point and a bit of the code being fucked with/deleted for some reason, but i was able to fix it

soap bar

21,084 views • 3 months ago

✨ I revived my first AI startup from 6 years ago with Claude Code [ 💡 ] Back then it used GPT-3 (this was 2 years before ChatGPT existed!) to generate new startup ideas which then people can vote on And the best startup ideas rise to the top! Back then I made it because people complained they didn't have any ideas to build a startup This week I moved it to its own VPS and installed Claude Code and told it to fix everything, the DB had become big and there was stupid write operations on every page load that it made it very slow Claude Code is excellent at fixing all those small bugs from old projects and quickly fixing them As Garry Tan says "boil the oceans" as in before I'd not have the time to fix these kinds of projects, it wouldn't be worth it, I mean IdeasAI doesn't even make money, but now it takes me an hour to do this and it works again! I also upgraded GPT-3 to xAI's Grok 4.2 for new startup ideas

✨ I revived my first AI startup from 6 years ago with Claude Code [ 💡 ] Back then it used GPT-3 (this was 2 years before ChatGPT existed!) to generate new startup ideas which then people can vote on And the best startup ideas rise to the top! Back then I made it because people complained they didn't have any ideas to build a startup This week I moved it to its own VPS and installed Claude Code and told it to fix everything, the DB had become big and there was stupid write operations on every page load that it made it very slow Claude Code is excellent at fixing all those small bugs from old projects and quickly fixing them As Garry Tan says "boil the oceans" as in before I'd not have the time to fix these kinds of projects, it wouldn't be worth it, I mean IdeasAI doesn't even make money, but now it takes me an hour to do this and it works again! I also upgraded GPT-3 to xAI's Grok 4.2 for new startup ideas

@levelsio

180,039 views • 5 months ago

✍️ I wrote an article on programming blur shaders for beginners. You do not need to be adept at either math or programming to understand how it works :) Going from the most basic/straightforward type of blur, to what you see below, and more! Lots of pretty pictures, media, and interactive code to learn from. -- Box blur, with clipping and radial weighting. (So, no longer really a box blur...?) 🔍 #gamedev #indiedev #unity3d

✍️ I wrote an article on programming blur shaders for beginners. You do not need to be adept at either math or programming to understand how it works :) Going from the most basic/straightforward type of blur, to what you see below, and more! Lots of pretty pictures, media, and interactive code to learn from. -- Box blur, with clipping and radial weighting. (So, no longer really a box blur...?) 🔍 #gamedev #indiedev #unity3d

Mirza Beig

121,190 views • 6 months ago