4 citations · 10 across the 11 of their papers we have counts for
9 papers · 1 filter
Robust Langevin Sampling under State-Dependent Heavy-Tailed Noise with Applications to LLM Fine-Tuning
M. Berk Sahin, Ege C. Kaya, Arda Fazla +2
Existing convergence guarantees for non-log-concave sampling with stochastic gradient Langevin Monte Carlo (SG-LMC) typically assume globally Lipschitz gradients and uniformly boun…
Why Does Adaptive Batching Help LLM Pretraining? A Perspective from Unbounded Variance
Arda Fazla, Antesh Upadhyay, Ege C. Kaya +2
Increasing the batch size during training is a common practice in large language model (LLM) pretraining, yet the theoretical justification behind its success is not well understoo…
EGGROLL, Unrolled: Understanding and Improving Low-Rank Evolution Strategies at Scale
Ege C. Kaya, Abolfazl Hashemi
EGGROLL (Sarkar et al., 2026) makes evolution strategies (ES) practical for LLMs by replacing dense Gaussian weight perturbations with low-rank Gaussian products, often of rank one…
Learning to Control Coupled-Dynamics Environments with Joint Markov Decision Processes
Ege C. Kaya, Aliasghar Pourghani, Mahsa Ghasemi +2
Coupled-dynamics environments expose the one-step outcomes that would follow from several possible counterfactual actions under a common realization of exogenous randomness. The or…
A Banach-Space Theory of Markovian Halpern Iteration for Non-Expansive Maps
Ege C. Kaya, Arda Fazla, M. Berk Sahin +1
We study stochastic approximation of fixed points of a non-expansive operator when the oracle samples originate from a continuing Markovian trajectory. A direct block-minibatch…
Quotient-Categorical Representations for Bellman-Compatible Average-Reward Distributional Reinforcement Learning
Ege C. Kaya, Aliasghar Pourghani, Vijay Gupta +1
Average-reward reinforcement learning requires estimating the gain and the bias, which is defined only up to an additive constant. This makes direct distributional analogues ill-po…
A Sharp Finite-Iteration Theory for Asynchronous Categorical Distributional Temporal-Difference Learning
Ege C. Kaya, Abolfazl Hashemi
We study finite-iteration behavior of asynchronous categorical distributional temporal-difference methods, covering scalar categorical TD (CTD) in the Cramér geometry and multivari…
Lower Bounds and Proximally Anchored SGD for Non-Convex Minimization Under Unbounded Variance
Arda Fazla, Ege C. Kaya, Antesh Upadhyay +1
Analysis of Stochastic Gradient Descent (SGD) and its variants typically relies on the assumption of uniformly bounded variance, a condition that frequently fails in practical non-…
Joint MDPs and Reinforcement Learning in Coupled-Dynamics Environments
Ege C. Kaya, Mahsa Ghasemi, Abolfazl Hashemi
Many distributional quantities in reinforcement learning are intrinsically joint across actions, including distributions of gaps and probabilities of superiority. However, the clas…