Research map / Reinforcement learning / Policy gradient and actor-critic
Policy and Value Transfer in Lifelong Reinforcement Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 the average-MDP policy instead of the action-prior policy as the jumpstart policy | jumpstart policy in Lava World (R~D) · Lava World | cumulative reward | Helped |
| Use the average-MDP policy instead of the action-prior policy as the jumpstart policy | jumpstart policy in Octogrid (G~D) · Octogrid | cumulative reward | Hurt |
| Initialize Q-learning with MAXQINIT instead of VMAX or average-MDP Q* | Q-learning in Tight and Spread grids · Multitask Tight/Spread Grid 11x11 | average reward | Helped |
| Initialize R-Max and Delayed-Q with MAXQINIT instead of VMAX | R-Max and Delayed-Q · Multitask Tight/Spread Grid 11x11 | average reward | Helped |
| Use a small number of sampled tasks (k=20 vs 60) to build MAXQINIT estimate (ablation) | MAXQINIT · 11x11 grid, 60 tasks | average reward | No clear effect |
Closest papers by meaning
- Lipschitz Lifelong Reinforcement Learning (AAAI 2021 · similarity 0.91)
- State Abstractions for Lifelong Reinforcement Learning (ICML 2018 · similarity 0.90)
- Model-based Lifelong Reinforcement Learning with Bayesian Exploration (NeurIPS 2022 · similarity 0.90)
- Oracle-Efficient Reinforcement Learning for Max Value Ensembles (NeurIPS 2024 · similarity 0.88)
- Fast TRAC: A Parameter-Free Optimizer for Lifelong Reinforcement Learning (NeurIPS 2024 · similarity 0.88)
- Long-Horizon Q-Learning: Accurate Value Learning via n-Step Inequalities (NeurIPS 2026 · 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.