Загрузка видео...

Не удалось загрузить видео

На главную

Stein Variational Gradient Descent (SVGD) is a particle-based approach to variational inference. SVGD iteratively transports a set of particles to match a target distribution by applying a form of functional gradient descent that minimizes the KL divergence.

alec_helbling's profile picture

Alec Helbling

10,798 subscribers

27,810 просмотров • 1 месяц назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Langevin Monte Carlo allows you to draw samples from a probability distribution using its log gradient ∇ log p(x). By performing a sort of gradient ascent with noise you can navigate around the distribution. Langevin MC is heavily related to modern diffusion models.
0:19

Langevin Monte Carlo allows you to draw samples from a probability distribution using its log gradient ∇ log p(x). By performing a sort of gradient ascent with noise you can navigate around the distribution. Langevin MC is heavily related to modern diffusion models.

Alec Helbling

96,895 просмотров • 1 год назад

Oldies but goldies: Y Nesterov, A method for solving a convex programming problem with convergence rate O(1/k^2). Improves the 1/k rate of vanilla gradient descent to 1/k^2.
0:10

Oldies but goldies: Y Nesterov, A method for solving a convex programming problem with convergence rate O(1/k^2). Improves the 1/k rate of vanilla gradient descent to 1/k^2.

Gabriel Peyré

47,616 просмотров • 1 год назад

Oldies but goldies: B Poljak, Some methods of speeding up the convergence of iteration methods, 1964. Introduces momentum methods to speed up gradient descent schemes.
0:10

Oldies but goldies: B Poljak, Some methods of speeding up the convergence of iteration methods, 1964. Introduces momentum methods to speed up gradient descent schemes.

Gabriel Peyré

27,859 просмотров • 1 год назад

Neat li'l #CSS gradient trick: A gradient circular chart effect 👀 1. a gradient mask 2. a conic gradient all the details on my blog
0:16

Neat li'l #CSS gradient trick: A gradient circular chart effect 👀 1. a gradient mask 2. a conic gradient all the details on my blog

Adam Argyle

67,503 просмотров • 3 лет назад

Congratulations also to Patrick 👍 for his #ICLR paper on Temporal Difference (TD) learning , in it We solve the decades-old puzzle of why TD can solve complex RL tasks that Gradient Descent cannot.
0:10

Congratulations also to Patrick 👍 for his #ICLR paper on Temporal Difference (TD) learning , in it We solve the decades-old puzzle of why TD can solve complex RL tasks that Gradient Descent cannot.

Thuerey Group at TUM

22,074 просмотров • 1 год назад

Flow Matching aims to learn a "flow" that transforms a simple source distribution (e.g. Gaussian) to an arbitrarily complex target distribution. This video shows the evolution of the marginal probability path as a source distribution is transformed to a target distribution.
0:13

Flow Matching aims to learn a "flow" that transforms a simple source distribution (e.g. Gaussian) to an arbitrarily complex target distribution. This video shows the evolution of the marginal probability path as a source distribution is transformed to a target distribution.

Alec Helbling

150,686 просмотров • 1 год назад

When optimization problems have multiple minima, algorithms favor specific solutions due to their implicit bias. For ordinary least squares (OLS), gradient descent inherently converges to the minimal norm solution among all possible solutions.
0:10

When optimization problems have multiple minima, algorithms favor specific solutions due to their implicit bias. For ordinary least squares (OLS), gradient descent inherently converges to the minimal norm solution among all possible solutions.

Samuel Vaiter

20,432 просмотров • 1 год назад

“The descent of Ganga is the descent of grace, guided by Shiva.” Maghi Purnima: Today Divine darshan of Ma Ganga appearing on your timeline💦
0:17

“The descent of Ganga is the descent of grace, guided by Shiva.” Maghi Purnima: Today Divine darshan of Ma Ganga appearing on your timeline💦

Jaya_Upadhyaya

73,850 просмотров • 5 месяцев назад

That gradient shift is wilddd. Insane attention to detail by Oğuz Yağız Kara
0:12

That gradient shift is wilddd. Insane attention to detail by Oğuz Yağız Kara

Ali Grids

22,897 просмотров • 5 месяцев назад

Drones followed by a wave of missiles. Tehran is applying the drone approach to disperse air defenses and facilitate hypersonic missile wave missions.
0:17

Drones followed by a wave of missiles. Tehran is applying the drone approach to disperse air defenses and facilitate hypersonic missile wave missions.

Sprinter Press Agency

528,841 просмотров • 1 год назад

[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.
0:11

[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

48,348 просмотров • 1 год назад

In 2023, neuroscientists uncovered that our brain distinguishes reality from imagination on a gradient, with a threshold marking the jump from one state to another. Learn about the biological breakthroughs of 2023:
0:11

In 2023, neuroscientists uncovered that our brain distinguishes reality from imagination on a gradient, with a threshold marking the jump from one state to another. Learn about the biological breakthroughs of 2023:

Quanta Magazine

74,292 просмотров • 2 лет назад

I took album cover of GNX by Kendrick Lamar and turned it into a 3D particle model in Threejs with TSL Used Grok 4 by xAI to improve the noise that moves particles.
0:16

I took album cover of GNX by Kendrick Lamar and turned it into a 3D particle model in Threejs with TSL Used Grok 4 by xAI to improve the noise that moves particles.

Dominik Fojcik

189,301 просмотров • 11 месяцев назад

A leaf with ink moves across water due to the Marangoni effect, where the ink's lower surface tension creates a gradient that propels the leaf forward.
0:13

A leaf with ink moves across water due to the Marangoni effect, where the ink's lower surface tension creates a gradient that propels the leaf forward.

Science girl

1,498,454 просмотров • 10 месяцев назад

Just a 🇻🇪 Venezuelan girl with a 🇨🇴 Colombian passport by descent residing in 🇦🇷 Argentina approved to go to the 🇺🇸 USA LEGALLY.
0:17

Just a 🇻🇪 Venezuelan girl with a 🇨🇴 Colombian passport by descent residing in 🇦🇷 Argentina approved to go to the 🇺🇸 USA LEGALLY.

Gaby💙

148,527 просмотров • 1 год назад

First video Witkoff (of Russian descent) meeting Putin (another Russian) and second video of Witkoff (of Russian descent) meeting Ukrainians. That`s all you need to see.
0:18

First video Witkoff (of Russian descent) meeting Putin (another Russian) and second video of Witkoff (of Russian descent) meeting Ukrainians. That`s all you need to see.

raging545

174,429 просмотров • 7 месяцев назад