Picture for Karl Hajjar

Karl Hajjar

LMO, CELESTE

On the symmetries in the dynamics of wide two-layer neural networks

Add code
Dec 06, 2022
Figure 1 for On the symmetries in the dynamics of wide two-layer neural networks
Figure 2 for On the symmetries in the dynamics of wide two-layer neural networks
Figure 3 for On the symmetries in the dynamics of wide two-layer neural networks
Viaarxiv icon

Training Integrable Parameterizations of Deep Neural Networks in the Infinite-Width Limit

Add code
Oct 29, 2021
Figure 1 for Training Integrable Parameterizations of Deep Neural Networks in the Infinite-Width Limit
Figure 2 for Training Integrable Parameterizations of Deep Neural Networks in the Infinite-Width Limit
Figure 3 for Training Integrable Parameterizations of Deep Neural Networks in the Infinite-Width Limit
Figure 4 for Training Integrable Parameterizations of Deep Neural Networks in the Infinite-Width Limit
Viaarxiv icon

SizeFlags: Reducing Size and Fit Related Returns in Fashion E-Commerce

Add code
Jun 07, 2021
Figure 1 for SizeFlags: Reducing Size and Fit Related Returns in Fashion E-Commerce
Figure 2 for SizeFlags: Reducing Size and Fit Related Returns in Fashion E-Commerce
Figure 3 for SizeFlags: Reducing Size and Fit Related Returns in Fashion E-Commerce
Figure 4 for SizeFlags: Reducing Size and Fit Related Returns in Fashion E-Commerce
Viaarxiv icon

Ranked Reward: Enabling Self-Play Reinforcement Learning for Combinatorial Optimization

Add code
Jul 06, 2018
Figure 1 for Ranked Reward: Enabling Self-Play Reinforcement Learning for Combinatorial Optimization
Figure 2 for Ranked Reward: Enabling Self-Play Reinforcement Learning for Combinatorial Optimization
Figure 3 for Ranked Reward: Enabling Self-Play Reinforcement Learning for Combinatorial Optimization
Figure 4 for Ranked Reward: Enabling Self-Play Reinforcement Learning for Combinatorial Optimization
Viaarxiv icon