3 papers
cs.LG2026
FIRM: Federated In-client Regularized Multi-objective Alignment for Large Language Models
Fatemeh Nourzad, Amirhossein Roknilamouki, Eylem Ekici +2
Aligning Large Language Models (LLMs) with human values often involves balancing multiple, conflicting objectives such as helpfulness and harmlessness. Training these models is com…
cs.LG2026
Escaping Offline Pessimism: Vector-Field Reward Shaping for Safe Frontier Exploration
Amirhossein Roknilamouki, Arnob Ghosh, Eylem Ekici +1
While offline reinforcement learning provides reliable policies for real-world deployment, its inherent pessimism severely restricts an agent's ability to explore and collect novel…
cs.LG2025
Provably Efficient RL for Linear MDPs under Instantaneous Safety Constraints in Non-Convex Feature Spaces
Amirhossein Roknilamouki, Arnob Ghosh, Ming Shi +3
In Reinforcement Learning (RL), tasks with instantaneous hard constraints present significant challenges, particularly when the decision space is non-convex or non-star-convex. Thi…