mlsanityData to 14 Oct 2026

Revealing the Learning Process in Reinforcement Learning Agents Through Attention-Oriented Metrics

arXiv 2406.14324 ↗First posted 20 Jun 2024cs.LGcited by 0

On the map: Deep reinforcement learning methods → Interpretable and logic-guided RL

Abstract (arXiv)

The learning process of a reinforcement learning (RL) agent remains poorly understood beyond the mathematical formulation of its learning algorithm. To address this gap, we introduce attention-oriented metrics (ATOMs) to investigate the development of an RL agent's attention during training. In a controlled experiment, we tested ATOMs on three variations of a Pong game, each designed to teach the agent distinct behaviours, complemented by a behavioural assessment. ATOMs successfully delineate the attention patterns of an agent trained on each game variation, and that these differences in attention patterns translate into differences in the agent's behaviour. Through continuous monitoring of ATOMs during training, we observed that the agent's attention developed in phases, and that these phases were consistent across game variations. Overall, we believe that ATOM could help improve our understanding of the learning processes of RL agents and better understand the relationship between attention and learning.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Interpretable and logic-guided RL, one of 15 directions in Deep reinforcement learning methods. Builds on 24 of its 37 references that are in the map. First posted on arXiv 20 Jun 2024.

Builds on

24 of 37 references.

References · 24

  1. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
  2. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  3. 2015Learning Deep Features for Discriminative Localizationcited by 2,436
  4. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  5. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  6. 2016Asynchronous Methods for Deep Reinforcement Learningcited by 2,258
  7. 2019Unmasking Clever Hans Predictors and Assessing What Machines Really Learncited by 365
  8. 2018Quantifying Generalization in Reinforcement Learningcited by 352
  9. 2015Explaining NonLinear Classification Decisions with Deep Taylor Decompositioncited by 250
  10. 2018Programmatically Interpretable Reinforcement Learningcited by 193
  11. 2019Explainable Reinforcement Learning Through a Causal Lenscited by 177
  12. 2017Visualizing and Understanding Atari Agentscited by 141
12 more
  1. 2020Convolutional Neural Networks as a Model of the Visual System: Past, Present, and Futurecited by 83
  2. 2017Hierarchical and Interpretable Skill Acquisition in Multi-task Reinforcement Learningcited by 77
  3. 2016Graying the black box: Understanding DQNscited by 68
  4. 2018Toward Interpretable Deep Reinforcement Learning with Linear Model U-Treescited by 65
  5. 2019Exploratory Not Explanatory: Counterfactual Analysis of Saliency Maps for Deep Reinforcement Learningcited by 60
  6. 2019Explain Your Move: Understanding Agent Actions Using Specific and Relevant Feature Attributioncited by 50
  7. 2018Contrastive Explanations for Reinforcement Learning in terms of Expected Consequencescited by 44
  8. 2019Conservative Q-Improvement: Reinforcement Learning for an Interpretable Decision-Tree Policycited by 29
  9. 2016Particle Swarm Optimization for Generating Interpretable Fuzzy Reinforcement Learning Policiescited by 19
  10. 2020Machine versus Human Attention in Deep Reinforcement Learning Taskscited by 16
  11. 2023A Dual-Stream Neural Network Explains the Functional Segregation of Dorsal and Ventral Visual Pathways in Human Brainscited by 12
  12. 2019Visual Rationalizations in Deep Reinforcement Learning for Atari Gamescited by 8

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Attention Trajectories as a Diagnostic Axis for Deep Reinforcement Learning
  2. 2025Interpretable Learning Dynamics in Unsupervised Reinforcement Learning
  3. 2025Revealing Human Attention Patterns from Gameplay Analysis for Reinforcement Learning
  4. 2024Attention-Driven Multi-Agent Reinforcement Learning: Enhancing Decisions with Expertise-Informed Tasks
  5. 2025Can you see how I learn? Human observers' inferences about Reinforcement Learning agents' learning processes
  6. 2026Understanding Goal Generalisation in Sequential Reinforcement Learning
  7. 2024Unraveling the Complexity of Memory in RL Agents: an Approach for Classification and Evaluation
  8. 2025Optimizing Sensory Neurons: Nonlinear Attention Mechanisms for Accelerated Convergence in Permutation-Invariant Neural Networks for Reinforcement Learning