Research map / Reinforcement learning / Policy gradient and actor-critic
Policy Gradient With Serial Markov Chain Reasoning
Edoardo Cetin, Oya Celiktutan
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 serial Markov chain reasoning with SSPG instead of traditional RL policies | Continuous-control agents · Mujoco suite | normalized performance | Improved |
| Use serial Markov chain reasoning with SSPG instead of baseline algorithms | Pixel-based control agents · DeepMind Control Suite | normalized performance | Improved |
| Increase the fixed number of reasoning steps (ablation) | SSPG · Two representative control tasks | performance | Improved |
| Use adaptive reasoning instead of a fixed number of reasoning steps (ablation) | SSPG · Two representative control tasks | performance | Improved |
| Initialize action beliefs from short-term action memory instead of a uniform distribution (ablation) | SSPG · Two control tasks | reasoning steps to convergence | Improved |
| Initialize action beliefs from short-term action memory instead of a uniform distribution (ablation) | SSPG · Two control tasks | performance | No clear effect |
Closest papers by meaning
- Bayesian Policy Gradient and Actor-Critic Algorithms (JMLR 2016 · similarity 0.88)
- RLVMR: Reinforcement Learning with Verifiable Meta-Reasoning Rewards for Robust Long-Horizon Agents (ICLR 2026 · similarity 0.88)
- Context-Sensitive Abstractions for Reinforcement Learning with Parameterized Actions (AAAI 2026 · similarity 0.88)
- Policy Gradient Methods in the Presence of Symmetries and State Abstractions (JMLR 2024 · similarity 0.88)
- Periodic agent-state based Q-learning for POMDPs (NeurIPS 2024 · similarity 0.88)
- Oracle-Efficient Reinforcement Learning for Max Value Ensembles (NeurIPS 2024 · 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.
