Research map / Reinforcement learning / Policy gradient and actor-critic
Test-driven Reinforcement Learning in Continuous Control
Zhao Yu, Xiuping Wu, Liangjun Ke
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 |
|---|---|---|---|
| learn rewards from test functions (TdRL) instead of using oracle handcrafted reward | SAC · DM-Control Walker-Stand, Walker-Run, Cheetah-Run, Quadruped-Run | episode return | No clear effect |
| replace SAC with PPO as policy optimizer in TdRL | TdRL · DM-Control tasks | episode return | Mixed |
| use pass-fail tests to define per-metric thresholds instead of weighted reward | TdRL · DM-Control Walker-Run | multi-objective satisfaction | Improved |
| trajectory-level test evaluation instead of state-action rewards | TdRL · DM-Control Walker-Run | stability | Hurt |
| add return penalty term L_Penalty (ablation) | TdRL · DM-Control (Figure 4a) | episode return | Improved |
| learn trajectory return first then decompose to reward, instead of direct reward learning (ablation) | TdRL · DM-Control (Figure 4a) | episode return | Improved |
| use a moderate multiple K_ES in early stop (not too large or small) (ablation) | TdRL-ES · DM-Control (Figure 4b) | episode return | Improved |
| TdRL-GN versus TdRL-ES variants of balancing the two losses (ablation) | TdRL · DM-Control tasks | episode return | No clear effect |
Closest papers by meaning
- R*: Efficient Reward Design via Reward Structure Evolution and Parameter Alignment Optimization with Large Language Models (ICML 2025 · similarity 0.88)
- TTRL: Test-Time Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- Efficient Test-Time Adaptation For Robot Policies (NeurIPS 2026 · similarity 0.87)
- Going Beyond Heuristics by Imposing Policy Improvement as a Constraint (NeurIPS 2024 · similarity 0.86)
- Efficient Language-instructed Skill Acquisition via Reward-Policy Co-Evolution (AAAI 2025 · similarity 0.86)
- TD-MPC2: Scalable, Robust World Models for Continuous Control (ICLR 2024 · similarity 0.86)
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.
