Research map / Reinforcement learning / MDPs and planning
Addressing Environment Non-Stationarity by Repeating Q-learning Updates
Sherief Abdallah, Michael Kaisers
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use RUQL instead of Q-learning | Q-learning · MAB switching (alpha=0.01, p0=p1=0.01, D=inf) | mean payoff: 0.69 -> 0.89 | Improved |
| use RUQL instead of Q-learning | Q-learning · MAB switching (alpha=0.01, D=500) | mean payoff: 0.58 -> 0.72 | Improved |
| use RUQL instead of FAQL | FAQL · MAB switching (alpha=0.01, D=250) | mean payoff: 0.57 -> 0.62 | Improved |
| use RUQL instead of Q-learning in stationary low-noise MAB, best parameters | Q-learning · MAB switching (D=inf, p0=p1=0.01) | mean payoff: 0.98 -> 0.97 | No clear effect |
| increase learning rate alpha from 0.01 to 0.1 (ablation) | Q-learning · MAB switching (p0=p1=0.01, D=inf) | mean payoff: 0.69 -> 0.09 | Hurt |
| use RUQL instead of Q-learning | Q-learning · 1-Row domain (P=0.1, D=5000) | mean payoff: 0.83 -> 0.90 | Improved |
| use Dyna-Q instead of Q-learning in noisy 1-Row | Q-learning · 1-Row domain (P=0.01, D=5000) | mean payoff: 0.74 -> -0.01 | Hurt |
| use Dyna-Q instead of Q-learning in noise-free 1-Row | Q-learning · 1-Row domain (P=0.0, D=5000) | mean payoff: 0.77 -> 1.31 | Improved |
| use RUQL instead of Q-learning | Q-learning · Taxi (p=0.01, D=inf) | mean payoff: -0.61 -> 0.45 | Improved |
| use RUQL instead of Q-learning | Q-learning · Taxi (p=0, D=300000) | mean payoff: 0.64 -> 1.19 | Improved |
| use RUQL instead of FAQL | FAQL · Taxi (p=0.01, D=300000) | mean payoff: -0.94 -> -0.70 | Improved |
| use RUQL instead of Q-learning | Q-learning · Mines (p=0.01, D=250000) | mean payoff: 0.27 -> 0.43 | Improved |
Closest papers by meaning
- Periodic agent-state based Q-learning for POMDPs (NeurIPS 2024 · similarity 0.89)
- Long-Horizon Q-Learning: Accurate Value Learning via n-Step Inequalities (NeurIPS 2026 · similarity 0.88)
- A Perspective of Q-value Estimation on Offline-to-Online Reinforcement Learning (AAAI 2024 · similarity 0.88)
- Pausing Policy Learning in Non-stationary Reinforcement Learning (ICML 2024 · similarity 0.88)
- Diffusion Policies Creating a Trust Region for Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.87)
- Adaptive $Q$-Network: On-the-fly Target Selection for Deep Reinforcement Learning (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
