Research map / Reinforcement learning / Visual and representation RL
Learning Retrospective Knowledge with Reverse Reinforcement Learning
Shangtong Zhang, Vivek Veeriah, Shimon Whiteson
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 |
|---|---|---|---|
| Add Reverse GVF auxiliary tasks to IMPALA | IMPALA · 10 Atari games | game performance: 7 out of 10 games; improvement is larger than 25% in 5 games | Improved |
| Use Reverse GVF instead of reward prediction as the auxiliary task | IMPALA · 10 Atari games | game performance: all 10 tested games | Improved |
| Use Reverse GVF instead of pixel control as the auxiliary task | IMPALA · 10 Atari games | game performance: 8 out of 10 tested games | Mixed |
| Use Reverse GVF rather than predictive GVF auxiliary tasks | IMPALA · 10 Atari games | game performance: 3 games | Mixed |
| Use smaller λ in Reverse TD(λ) (ablation) | Reverse TD(λ) · Microdrone | mean value error | Improved |
| Use a single observed reverse return and its state-conditioned learned distribution for anomaly detection | Distributional Reverse GVF · Microdrone anomaly detection | anomaly probability | Improved |
| Use a single observed reverse return and its state-conditioned learned distribution for anomaly detection | Distributional Reverse GVF · Reacher anomaly detection | probability of anomaly: higher than 0.8 | Mixed |
Closest papers by meaning
- Rewriting History with Inverse RL: Hindsight Inference for Policy Improvement (NeurIPS 2020 · similarity 0.85)
- Value-driven Hindsight Modelling (NeurIPS 2020 · similarity 0.84)
- ERCI: An Explainable Experience Replay Approach with Causal Inference for Deep Reinforcement Learning (AAAI 2025 · similarity 0.83)
- SAGE: Shaping Anchors for Guided Exploration in RLVR of LLMs (ICML 2026 · similarity 0.83)
- Generalization of RLVR Using Causal Reasoning as a Testbed (ICLR 2026 · similarity 0.83)
- Is Inverse Reinforcement Learning Harder than Standard Reinforcement Learning? A Theoretical Perspective (ICML 2024 · similarity 0.82)
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.
