Search

Egor Shulgin

Beyond the Ideal: Analyzing the Inexact Muon Update
Gluon: Making Muon & Scion Great Again! (Bridging Theory and Practice of LMO-based Optimizers for LLMs)
Smoothed Normalization for Efficient Distributed Private Optimization
MAST: Model-Agnostic Sparsified Training
On the Convergence of DP-SGD with Adaptive Clipping
Towards a Better Theoretical Understanding of Independent Subnetwork Training
Shifted Compression Framework: Generalizations and Improvements
ADOM: Accelerated Decentralized Optimization Method for Time-Varying Networks
Revisiting Stochastic Extragradient
Uncertainty Principle for Communication Compression in Distributed and Federated Learning and the Search for an Optimal Compressor
Adaptive Catalyst for Smooth Convex Optimization
SGD: General Analysis and Improved Rates
Lecture Notes on Stochastic Processes

Powered by the Academic theme for Hugo.