Research map / Reinforcement learning / Visual and representation RL
Toward Deep Reinforcement Learning Without a Simulator: An Autonomous Steering Example
Bar Hilleli, Ran El-Yaniv
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 |
|---|---|---|---|
| initialize Q-network from imitation-learned policy instead of random init (ablation) | DDQN driving agent · Assetto Corsa lane driving, epochs 0-3 | accidents per epoch: 116.6 -> 67 | Improved |
| initialize Q-network from imitation-learned policy instead of random init (ablation) | DDQN driving agent · Assetto Corsa lane driving, epochs 39-120 | accidents per epoch: 41 -> 44.5 | No clear effect |
| add safety module to IL-initialized RL (ablation) | DDQN driving agent with IL init · Assetto Corsa lane driving, epochs 0-3 | accidents per epoch: 67 -> 47 | Improved |
| add safety module to IL-initialized RL (ablation) | DDQN driving agent with IL init · Assetto Corsa lane driving, epochs 39-120 | accidents per epoch: 44.5 -> 23.3 | Improved |
| add safety module to randomly initialized RL (ablation) | DDQN driving agent with random init · Assetto Corsa lane driving | average reward per epoch | Improved |
| train reward network on full labeled data rather than 20% (ablation) | learned reward network · Assetto Corsa lane driving | RL agent reward: 66% -> 97% reward-net validation accuracy | Improved |
Closest papers by meaning
- CIRL: Controllable Imitative Reinforcement Learning for Vision-based Self-driving (ECCV 2018 · similarity 0.88)
- End-to-End Urban Driving by Imitating a Reinforcement Learning Coach (ICCV 2021 · similarity 0.87)
- RLfOLD: Reinforcement Learning from Online Demonstrations in Urban Autonomous Driving (AAAI 2024 · similarity 0.86)
- Prioritizing Perception-Guided Self-Supervision: A New Paradigm for Causal Modeling in End-to-End Autonomous Driving (NeurIPS 2025 · similarity 0.85)
- Learning to Drive in New Cities Without Human Demonstrations (NeurIPS 2026 · similarity 0.85)
- Distill to Think, Foresee to Act: Cognitive-Physical Reinforcement Learning for Autonomous Driving (NeurIPS 2026 · similarity 0.85)
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.
