Research map / Reinforcement learning / Policy gradient and actor-critic
Q-Pensieve: Boosting Sample Efficiency of Multi-Objective RL Through Memory Sharing of Q-Snapshots
Wei Hung, Bo Kai Huang, Ping-Chun Hsieh, Xi Liu
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 |
|---|---|---|---|
| Q-Pensieve (Q replay buffer size 4, 5-preference set) vs explicit search baselines and CN-DER | Q-Pensieve · 10 domains, Table 1 | HV, UT, ED | Improved |
| Q replay buffer (Q-snapshots from past iterations) vs Vanilla without buffer (ablation) | Q-Pensieve · Ablation study, DST/HalfCheetah | HV and scalarized return | Improved |
| Hybrid SAC switching to Q-Pensieve update using Q-snapshots from a parallel Q-Pensieve model at 1e5 steps (ablation) | single-objective SAC · DST and HalfCheetah | return per preference | Improved |
| Q-Pensieve vs 19 single-objective SAC models on 2-objective MuJoCo | Q-Pensieve · 2-objective MuJoCo, 19 preferences | return vectors vs SAC collection: 1.5M total steps vs 28.5M for SAC | Improved |
| Explicit search baselines (PFA, PGMORL) given 10x samples | PFA, PGMORL · Table 1 domains | HV, UT, ED | Hurt |
| Explicit search (PFA/PGMORL) vs implicit search (CN-DER) | MORL method family · Table 1 | HV vs UT | Mixed |
Later papers that compared with this method
1 later accepted paper reports Q-Pensieve as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Multi-Objective Reinforcement Learning Using Routed Ensembles and Trajectory Attribution (NeurIPS 2026 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Finite-Time Convergence and Sample Complexity of Actor-Critic Multi-Objective Reinforcement Learning (ICML 2024 · similarity 0.88)
- Pareto Set Learning for Multi-Objective Reinforcement Learning (AAAI 2025 · similarity 0.87)
- On Generalization Across Environments In Multi-Objective Reinforcement Learning (ICLR 2025 · similarity 0.87)
- Preference Conditioned Multi-Objective Reinforcement Learning: Decomposed, Diversity-Driven Policy Optimization (NeurIPS 2026 · similarity 0.87)
- Diffusion Policies Creating a Trust Region for Offline Reinforcement Learning (NeurIPS 2024 · similarity 0.87)
- QMP: Q-switch Mixture of Policies for Multi-Task Behavior Sharing (ICLR 2025 · 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.
