AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Guided Exploration with Proximal Policy Optimization using a Single Demonstration

Gabriele Libardi, Gianni De Fabritiis, Sebastian Dittert

ICML 2021 · Reinforcement learning › Visual and representation RL · approach: Novelty and impact-driven exploration · 2 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use PPO+D with one demonstration instead of the tested imitation-learning and PPO baselinesAnimal-AI Olympics hard-exploration tasks · One box easy, One box hard, Two boxes easy, and Two boxes hardrewardImproved
Add value-prioritized replay of unsuccessful trajectories (ablation)PPO+D · Two boxes easyrewardImproved
Remove value-buffer replay (ablation)PPO+D · One box easylearning speedMixed
Remove value-buffer replay (ablation)PPO+D · One box easytraining reliabilityHurt
Use PPO+D instead of vanilla PPOSparse ReacherPyBulletEnv-v0 · Sparse ReacherPyBulletEnv-v0rewardImproved
Use PPO+D instead of PPO+BC, with or without a successful-trajectory bufferSparse ReacherPyBulletEnv-v0 · Sparse ReacherPyBulletEnv-v0rewardImproved
Use PPO+D instead of PPO+BCSparse LunarLander-v2 · Sparse LunarLander-v2rewardImproved
Use PPO+D instead of vanilla PPOSparse LunarLander-v2 · Sparse LunarLander-v2rewardMixed
Apply PPO+D to a dense-reward task instead of vanilla PPOPPO+D · BreakoutNoFrameskip-v4learning speedHurt
Set the demonstration replay probability to rho=0.3 rather than the other tested values (ablation)PPO+D with phi=0 · One box easy, fixed agent positionperformanceImproved
Set demonstration replay probability too low (ablation)PPO+D · One box easy, fixed agent positionperformanceHurt
Set demonstration replay probability too high (ablation)PPO+D · One box easy, fixed agent positionperformanceHurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy