A selection of research publications and presentations.
Conference Paper2026
Empirical Evaluation of Policy-Gradient Reinforcement Learning for Dynamic Service Control in an M/M/1 Queue
Joseph A. Walton, Gabriel Nicolosi
IISE Annual Conference & Expo · Presented
Abstract
This work evaluates policy-gradient reinforcement learning methods for service-rate control in a stochastic queueing system, with attention to policy structure, training stability, and simulation-based validation.
Working Paper2026
State-Action Space Growth in Discrete Machine-Interference Repair Control with Speed-Dependent Failures
Joseph A. Walton, Gabriel Nicolosi
Working Paper · In preparation
Abstract
A study of state and action growth, exact dynamic programming boundaries, and scalable solution approaches for controlled finite-source machine repair systems.
Poster2025
Simulation-Based Evaluation of Reinforcement Learning Policies for Dynamic Queue Control
Joseph A. Walton, Gabriel Nicolosi
Research Poster · Presented
Abstract
A visual summary of reinforcement learning experiments for adaptive queueing control under uncertain arrivals and operating costs.
