Research map / Reinforcement learning / Visual and representation RL
RIDE: Rewarding Impact-Driven Exploration for Procedurally-Generated Environments
Roberta Raileanu, Tim Rocktäschel
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 |
|---|---|---|---|
| Use impact-driven intrinsic reward with episodic state-count discounting | IMPALA on procedurally generated MiniGrid · MiniGrid hard exploration tasks | sample efficiency and task success | Improved |
| Use impact-driven reward rather than prediction-error curiosity | RIDE versus ICM · Mario, first level | learning performance | Improved |
| Use RIDE exploration | IMPALA on VizDoom · VizDoom | task performance | Improved |
| Use impact-driven reward with episodic state-count discounting (ablation) | RIDE · MiniGrid hard exploration tasks | task success | Improved |
| Add impact-driven reward to episodic state counts (ablation) | OnlyEpisodicCounts · MiniGrid hard exploration tasks | task success | Improved |
| Retain entropy regularization (ablation) | NoEpisodicCounts RIDE variant · MiniGrid hard exploration tasks | task success | Improved |
| Use learned state embeddings instead of raw observations to compute impact (ablation) | RIDE · MiniGrid tasks | task success | Improved |
| Keep policy and impact embedding representations separate (ablation) | RIDE · MiniGrid tasks | task success | Improved |
| Add a noisy, randomly recoloring TV distractor | ICM · MiniGridN7S4NoisyTV | task performance | Hurt |
| Train with intrinsic and extrinsic rewards | ICM · Mario, first level | learning performance | Hurt |
Later papers that compared with this method
5 later accepted papers report RIDE as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- Rethinking Exploration in Reinforcement Learning with Effective Metric-Based Exploration Bonus (NeurIPS 2024 · better on 1 of 4 shared results)
- Efficient Potential-based Exploration in Reinforcement Learning using Inverse Dynamic Bisimulation Metric (NeurIPS 2023 · better on 7 of 7 shared results)
- EAGER: Asking and Answering Questions for Automatic Reward Shaping in Language-guided RL (NeurIPS 2022 · better on 1 of 1 shared result)
- Adversarially Guided Actor-Critic (ICLR 2021 · better on 6 of 8 shared results)
- Learning with AMIGo: Adversarially Motivated Intrinsic Goals (ICLR 2021 · 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
- DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning (NeurIPS 2025 · similarity 0.89)
- Decoupling Exploration and Policy Optimization: Uncertainty Guided Tree Search for Hard Exploration (NeurIPS 2026 · similarity 0.89)
- Action-Dependent Optimality-Preserving Reward Shaping (ICML 2025 · similarity 0.88)
- CIG: Exploration via Conditional Information Gain (NeurIPS 2026 · similarity 0.87)
- Abstract and Explore: A Novel Behavioral Metric with Cyclic Dynamics in Reinforcement Learning (AAAI 2024 · similarity 0.87)
- Efficient Exploration and Discriminative World Model Learning with an Object-Centric Abstraction (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.
