Pinned Loading

1. polargrad polargrad Public

   PolarGrad: A Class of Matrix-Gradient Optimizers from a Unifying Preconditioning Perspective

   Python 18

2. BCD-for-DNNs-PyTorch BCD-for-DNNs-PyTorch Public

   Code for Global Convergence of Block Coordinate Descent in Deep Learning (ICML 2019)

   Jupyter Notebook 37 8

3. bregman_prox_langevin_mc bregman_prox_langevin_mc Public

   Code for Bregman Proximal Langevin Monte Carlo via Bregman--Moreau Envelopes (ICML 2022)

   Jupyter Notebook 1 1

4. equivariant_optimizers equivariant_optimizers Public

   Code for Symmetry-Compatible Principle for Optimizer Design: Embeddings, LM Heads, SwiGLU MLPs, and MoE Routers

   Python 4