collaborators

7 papers

cs.LG2026

Adaptive Batch Sizes Using Non-Euclidean Gradient Noise Scales for Stochastic Sign and Spectral Descent

Hiroki Naganuma, Shagun Gupta, Youssef Briki +4

To maximize hardware utilization, modern machine learning systems typically employ large constant or manually tuned batch size schedules, relying on heuristics that are brittle and…

cs.LG2026

How to Allocate, How to Learn? Dynamic Rollout Allocation and Advantage Modulation for Policy Optimization

Yangyi Fang, Jiaye Lin, Xiaoliang Fu +6

Reinforcement Learning with Verifiable Rewards (RLVR) has proven effective for Large Language Model (LLM) reasoning, yet current methods face key challenges in resource allocation…

cs.LG2026

Smoothing DiLoCo with Primal Averaging for Faster Training of LLMs

Aaron Defazio, Konstantin Mishchenko, Parameswaran Raman +2

We propose Generalized Primal Averaging (GPA), an extension of Nesterov's method that unifies and generalizes recent averaging-based optimizers like single-worker DiLoCo and Schedu…

cs.AI2026

Proximity-Based Multi-Turn Optimization: Practical Credit Assignment for LLM Agent Training

Yangyi Fang, Jiaye Lin, Xiaoliang Fu +4

Multi-turn LLM agents are becoming pivotal to production systems, spanning customer service automation, e-commerce assistance, and interactive task management, where accurately dis…

cs.LG2026

Clarifying Shampoo: Adapting Spectral Descent to Stochasticity and the Parameter Trajectory

Runa Eschenhagen, Anna Cai, Tsung-Hsien Lee +1

Optimizers leveraging the matrix structure in neural networks, such as Shampoo and Muon, are more data-efficient than element-wise algorithms like Adam and Signum. While in specifi…

cs.LG2025

Purifying Shampoo: Investigating Shampoo's Heuristics by Decomposing its Preconditioner

Runa Eschenhagen, Aaron Defazio, Tsung-Hsien Lee +2

The recent success of Shampoo in the AlgoPerf contest has sparked renewed interest in Kronecker-factorization-based optimization algorithms for training neural networks. Despite it…