mlsanityData to 7 Oct 2026

Pay Attention to What and Where? Interpretable Feature Extractor in Vision-based Deep Reinforcement Learning

arXiv 2504.10071 ↗First posted 14 Apr 2025cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Current approaches in Explainable Deep Reinforcement Learning have limitations in which the attention mask has a displacement with the objects in visual input. This work addresses a spatial problem within traditional Convolutional Neural Networks (CNNs). We propose the Interpretable Feature Extractor (IFE) architecture, aimed at generating an accurate attention mask to illustrate both "what" and "where" the agent concentrates on in the spatial domain. Our design incorporates a Human-Understandable Encoding module to generate a fully interpretable attention mask, followed by an Agent-Friendly Encoding module to enhance the agent's learning efficiency. These two components together form the Interpretable Feature Extractor for vision-based deep reinforcement learning to enable the model's interpretability. The resulting attention mask is consistent, highly understandable by humans, accurate in spatial dimension, and effectively highlights important objects or locations in visual input. The Interpretable Feature Extractor is integrated into the Fast and Data-efficient Rainbow framework, and evaluated on 57 ATARI games to show the effectiveness of the proposed approach on Spatial Preservation, Interpretability, and Data-efficiency. Finally, we showcase the versatility of our approach by incorporating the IFE into the Asynchronous Advantage Actor-Critic Model.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 31 of its 42 references that are in the map. First posted on arXiv 14 Apr 2025.

Builds on

31 of 42 references.

References · 31

  1. 2017Attention Is All You Needcited by 59,932
  2. 2017Proximal Policy Optimization Algorithmscited by 14,441
  3. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  4. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,530
  5. 2018CBAM: Convolutional Block Attention Modulecited by 2,463
  6. 2016Asynchronous Methods for Deep Reinforcement Learningcited by 2,218
  7. 2015Deep Reinforcement Learning with Double Q-learningcited by 1,704
  8. 2015Show, Attend and Tell: Neural Image Caption Generation with Visual Attentioncited by 1,421
  9. 2018Experience Replay for Continual Learningcited by 1,074
  10. 2015Teaching Machines to Read and Comprehendcited by 1,039
  11. 2018IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architecturescited by 866
  12. 2017Rainbow: Combining Improvements in Deep Reinforcement Learningcited by 774
19 more
  1. 2014Recurrent Models of Visual Attentioncited by 344
  2. 2019Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Modelcited by 228
  3. 2018Compositional Attention Networks for Machine Reasoningcited by 183
  4. 2017Visualizing and Understanding Atari Agentscited by 141
  5. 2023A Definition of Continual Reinforcement Learningcited by 98
  6. 2019Towards Interpretable Reinforcement Learning Using Attention Augmented Agentscited by 82
  7. 2019Exploratory Not Explanatory: Counterfactual Analysis of Saliency Maps for Deep Reinforcement Learningcited by 60
  8. 2019Explain Your Move: Understanding Agent Actions Using Specific and Relevant Feature Attributioncited by 50
  9. 2021CORA: Benchmarks, Baselines, and Metrics as a Platform for Continual Reinforcement Learning Agentscited by 36
  10. 2022Disentangling Transfer in Continual Reinforcement Learningcited by 31
  11. 2022Human-level Atari 200x fastercited by 28
  12. 2015Deep Attention Recurrent Q-Networkcited by 27
  13. 2019Reinforcement Learning with Attention that Works: A Self-Supervised Approachcited by 17
  14. 2019Gossip-based Actor-Learner Architectures for Deep Reinforcement Learningcited by 16
  15. 2020Contrastive Explanations for Reinforcement Learning via Embedded Self Predictionscited by 13
  16. 2017Multi-focus Attention Network for Efficient Deep Reinforcement Learningcited by 10
  17. 2020Attentional Bottleneck: Towards an Interpretable Deep Driving Networkcited by 10
  18. 2021Fast and Data-Efficient Training of Rainbow: an Experimental Study on Ataricited by 10
  19. 2017Hierarchical Attentive Recurrent Trackingcited by 7

Built on by

Not cited yet.