Research map / Reinforcement learning / Visual and representation RL
Self-Supervised Attention-Aware Reinforcement Learning
Haiping Wu, Khimya Khetarpal, Doina Precup
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 self-supervised attention masks to gate CNN features | A2C and ACKTR on Atari games · Atari ALE single-task games | test score | Improved |
| Train a universal attention module jointly on Asteroids, Assault, and Ms. Pacman | A2C · Asteroids, Assault, and Ms. Pacman | test score | Improved |
| Transfer a fixed attention module trained on another game | A2C · Asteroids and Carnival | test score | Improved |
| Replace self-supervised attention with RL-objective-trained top-down attention | A2C · Asteroids, Carnival, Freeway, and Ms. Pacman | test score | Mixed |
| Use the proposed attention masks instead of Transporter keypoints | Object keypoint extraction · Frostbite | Recall: 0.74 -> 0.76 | Improved |
| Use the proposed attention masks instead of Transporter keypoints | Object keypoint extraction · Frostbite | Precision: 0.56 -> 0.57 | Improved |
| Use the proposed attention masks instead of Transporter keypoints | Object keypoint extraction · Berzerk | Recall: 0.32 -> 0.54 | Improved |
| Use the proposed attention masks instead of Transporter keypoints | Object keypoint extraction · Berzerk | Precision: 0.27 -> 0.46 | Improved |
| Use the proposed attention masks instead of Transporter keypoints | Object keypoint extraction · Ms. Pacman | Recall: 0.33 -> 0.72 | Improved |
| Use the proposed attention masks instead of Transporter keypoints | Object keypoint extraction · Ms. Pacman | Precision: 0.23 -> 0.52 | Improved |
Closest papers by meaning
- Attention Trajectories as a Diagnostic Axis for Deep Reinforcement Learning (NeurIPS 2026 · similarity 0.89)
- VLMs-Guided Representation Distillation for Efficient Vision-Based Reinforcement Learning (CVPR 2025 · similarity 0.88)
- Focus-Then-Decide: Segmentation-Assisted Reinforcement Learning (AAAI 2024 · similarity 0.87)
- 1000 Layer Networks for Self-Supervised RL: Scaling Depth Can Enable New Goal-Reaching Capabilities (NeurIPS 2025 · similarity 0.87)
- A Simple Framework for Generalization in Visual RL under Dynamic Scene Perturbations (NeurIPS 2024 · similarity 0.87)
- SiT: Symmetry-invariant Transformers for Generalisation in Reinforcement Learning (ICML 2024 · 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.
