4 papers
PriPG-RL: Privileged Planner-Guided Reinforcement Learning for Partially Observable Systems with Anytime-Feasible MPC
Mohsen Amiri, Ali Beikmohammadi, Sindri Magnuśson +1
This paper addresses the problem of training a reinforcement learning (RL) policy under partial observability by exploiting a privileged, anytime-feasible planner agent available e…
MARBLE: Multi-Armed Restless Bandits in Latent Markovian Environment
Mohsen Amiri, Konstantin Avrachenkov, Ibtihal El Mimouni +1
Restless Multi-Armed Bandits (RMABs) are powerful models for decision-making under uncertainty, yet classical formulations typically assume fixed dynamics, an assumption often viol…
Challenger-Based Combinatorial Bandits for Subcarrier Selection in OFDM Systems
Mohsen Amiri, V Venktesh, Sindri Magnússon
This paper investigates the identification of the top-m user-scheduling sets in multi-user MIMO downlink, which is cast as a combinatorial pure-exploration problem in stochastic li…
Reinforcement Learning in Switching Non-Stationary Markov Decision Processes: Algorithms and Convergence Analysis
Mohsen Amiri, Sindri Magnússon
We introduce the Switching Non-Stationary Markov Decision Process (SNS-MDP) framework, in which the environment transitions among a finite set of MDPs governed by a latent Markov c…