Research map / Reinforcement learning / Visual and representation RL
Constrained GPI for Zero-Shot Transfer in 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 |
|---|---|---|---|
| replace GPI with constrained GPI at test time (target task vector as C) | USFA agents · Scavenger d=2 | normalized score | Helped |
| replace GPI with constrained GPI at test time | USFA agents · Scavenger d=4 | normalized score | Helped |
| add constrained training to USFA with GPI | USFA agents · Scavenger d=4 | normalized score | Mixed |
| replace GPI with constrained GPI at test time | USFA with learned phi · DeepMind Lab left-to-right | score | Helped |
| replace GPI with constrained GPI at test time | USFA with learned phi · DeepMind Lab near-to-far | score | Helped |
| apply lower and upper bounds to action-values | Constrained GPI · Scavenger d=2 | greedy actions changed (%): 22.11% | Helped |
| use source task set as C instead of target task vector | GPI · Scavenger | normalized score | Hurt |
Closest papers by meaning
- Distributional Successor Features Enable Zero-Shot Policy Optimization (NeurIPS 2024 · similarity 0.87)
- SF-DQN: Provable Knowledge Transfer using Successor Feature for Deep Reinforcement Learning (ICML 2024 · similarity 0.87)
- Model-Based Transfer Learning for Contextual Reinforcement Learning (NeurIPS 2024 · similarity 0.87)
- Proto Successor Measure: Representing the Behavior Space of an RL Agent (ICML 2025 · similarity 0.86)
- How Ensembles of Distilled Policies Improve Generalisation in Reinforcement Learning (NeurIPS 2025 · similarity 0.86)
- Constructing an Optimal Behavior Basis for the Option Keyboard (NeurIPS 2025 · 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.