Alert button
Picture for Yatin Dandi

Yatin Dandi

Alert button

Asymptotics of feature learning in two-layer networks after one gradient-step

Feb 07, 2024
Hugo Cui, Luca Pesce, Yatin Dandi, Florent Krzakala, Yue M. Lu, Lenka Zdeborová, Bruno Loureiro

Viaarxiv icon

The Benefits of Reusing Batches for Gradient Descent in Two-Layer Networks: Breaking the Curse of Information and Leap Exponents

Feb 05, 2024
Yatin Dandi, Emanuele Troiani, Luca Arnaboldi, Luca Pesce, Lenka Zdeborová, Florent Krzakala

Viaarxiv icon

A Gentle Introduction to Gradient-Based Optimization and Variational Inequalities for Machine Learning

Sep 09, 2023
Neha S. Wadia, Yatin Dandi, Michael I. Jordan

Figure 1 for A Gentle Introduction to Gradient-Based Optimization and Variational Inequalities for Machine Learning
Figure 2 for A Gentle Introduction to Gradient-Based Optimization and Variational Inequalities for Machine Learning
Figure 3 for A Gentle Introduction to Gradient-Based Optimization and Variational Inequalities for Machine Learning
Figure 4 for A Gentle Introduction to Gradient-Based Optimization and Variational Inequalities for Machine Learning
Viaarxiv icon

Sampling with flows, diffusion and autoregressive neural networks: A spin-glass perspective

Aug 27, 2023
Davide Ghio, Yatin Dandi, Florent Krzakala, Lenka Zdeborová

Viaarxiv icon

Learning Two-Layer Neural Networks, One (Giant) Step at a Time

May 29, 2023
Yatin Dandi, Florent Krzakala, Bruno Loureiro, Luca Pesce, Ludovic Stephan

Figure 1 for Learning Two-Layer Neural Networks, One (Giant) Step at a Time
Figure 2 for Learning Two-Layer Neural Networks, One (Giant) Step at a Time
Figure 3 for Learning Two-Layer Neural Networks, One (Giant) Step at a Time
Figure 4 for Learning Two-Layer Neural Networks, One (Giant) Step at a Time
Viaarxiv icon

Universality laws for Gaussian mixtures in generalized linear models

Feb 17, 2023
Yatin Dandi, Ludovic Stephan, Florent Krzakala, Bruno Loureiro, Lenka Zdeborová

Figure 1 for Universality laws for Gaussian mixtures in generalized linear models
Figure 2 for Universality laws for Gaussian mixtures in generalized linear models
Figure 3 for Universality laws for Gaussian mixtures in generalized linear models
Figure 4 for Universality laws for Gaussian mixtures in generalized linear models
Viaarxiv icon

Data-heterogeneity-aware Mixing for Decentralized Learning

Apr 13, 2022
Yatin Dandi, Anastasia Koloskova, Martin Jaggi, Sebastian U. Stich

Figure 1 for Data-heterogeneity-aware Mixing for Decentralized Learning
Figure 2 for Data-heterogeneity-aware Mixing for Decentralized Learning
Figure 3 for Data-heterogeneity-aware Mixing for Decentralized Learning
Figure 4 for Data-heterogeneity-aware Mixing for Decentralized Learning
Viaarxiv icon

NeurInt : Learning to Interpolate through Neural ODEs

Nov 07, 2021
Avinandan Bose, Aniket Das, Yatin Dandi, Piyush Rai

Figure 1 for NeurInt : Learning to Interpolate through Neural ODEs
Figure 2 for NeurInt : Learning to Interpolate through Neural ODEs
Figure 3 for NeurInt : Learning to Interpolate through Neural ODEs
Figure 4 for NeurInt : Learning to Interpolate through Neural ODEs
Viaarxiv icon

Understanding Layer-wise Contributions in Deep Neural Networks through Spectral Analysis

Nov 06, 2021
Yatin Dandi, Arthur Jacot

Figure 1 for Understanding Layer-wise Contributions in Deep Neural Networks through Spectral Analysis
Figure 2 for Understanding Layer-wise Contributions in Deep Neural Networks through Spectral Analysis
Figure 3 for Understanding Layer-wise Contributions in Deep Neural Networks through Spectral Analysis
Figure 4 for Understanding Layer-wise Contributions in Deep Neural Networks through Spectral Analysis
Viaarxiv icon

Implicit Gradient Alignment in Distributed and Federated Learning

Jun 25, 2021
Yatin Dandi, Luis Barba, Martin Jaggi

Figure 1 for Implicit Gradient Alignment in Distributed and Federated Learning
Figure 2 for Implicit Gradient Alignment in Distributed and Federated Learning
Figure 3 for Implicit Gradient Alignment in Distributed and Federated Learning
Figure 4 for Implicit Gradient Alignment in Distributed and Federated Learning
Viaarxiv icon