Research map / Reinforcement learning / MDPs and planning
Analysis of approximate linear programming solution to Markov decision problem with log barrier function
Donghwan Lee, Hyukjun Yang, Bumgeun Park
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 |
|---|---|---|---|
| replace MSE Bellman critic loss with log-barrier LP loss in DDPG | DDPG · MuJoCo (Ant, Walker2d, HalfCheetah, Humanoid) | average return | Improved |
| replace MSE Bellman critic loss with log-barrier LP loss in DDPG | DDPG · MuJoCo Hopper | average return | No clear effect |
| replace MSE Bellman loss with log-barrier LP loss and no target network in DQN | DQN · Gymnasium control tasks (Acrobot, CartPole, LunarLander, MountainCar, Pendulum) | average return | No clear effect |
| replace MSE Bellman loss with log-barrier LP loss in DQN | DQN · CartPole-v1 | learning speed and stability | Improved |
| use alternative penalty such as SoftPlus instead of the log-barrier function (ablation) | Log-barrier DQN/DDPG · Gym/MuJoCo tasks | performance | Hurt |
| decrease barrier parameter eta (ablation) | log-barrier LP solution · tabular MDP (theory) | error norm ||Q_eta - Q*|| | Improved |
Closest papers by meaning
- Approximate Newton Methods for Policy Search in Markov Decision Processes (JMLR 2016 · similarity 0.88)
- Provably Efficient RL under Episode-Wise Safety in Constrained MDPs with Linear Function Approximation (NeurIPS 2025 · similarity 0.87)
- Revisiting the Linear-Programming Framework for Offline RL with General Function Approximation (ICML 2023 · similarity 0.87)
- Computationally Efficient RL under Linear Bellman Completeness for Deterministic Dynamics (ICLR 2025 · similarity 0.86)
- Solving Robust Markov Decision Processes: Generic, Reliable, Efficient (AAAI 2025 · similarity 0.85)
- Achieving $\tilde{O}(1/\epsilon)$ Sample Complexity for Constrained Markov Decision Process (NeurIPS 2024 · similarity 0.85)
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.
