Publications

2025

  1. SPEQ: Offline Stabilization Phases for Efficient Q-Learning in High Update-To-Data Ratio Reinforcement Learning
    Carlo Romeo, Girolamo Macaluso, Alessandro Sestini, and 1 more author
    In Reinforcement Learning Journal (RLJ), 2025
    To appear
  2. NTRL: Encounter Generation via Reinforcement Learning for Dynamic Difficulty Adjustment in Dungeons and Dragons
    Carlo Romeo and Andrew D. Bagdanov
    In IEEE Conference on Games (CoG), 2025
    To appear

2024

  1. Offline Reinforcement Learning with Imputed Rewards
    Carlo Romeo and Andrew D. Bagdanov
    2024
    In Reinforcement Learning Conference (RLC) @ RLBrew Workshop, 2024