Graying The Black Box: Understanding Dqns

Abstract

In recent years there is a growing interest in using deep representations for reinforcement learning. In this paper, we present a methodology and tools to analyze Deep Q-networks (DQNs) in a non-blind matter. Moreover, we propose a new model, the Semi Aggregated Markov Decision Process (SAMDP), and an algorithm that learns it automatically. The SAMDP model allows us to identify spatio-temporal abstractions directly from features and may be used as a sub-goal detector in future work. Using our tools we reveal that the features learned by DQNs aggregate the state space in a hierarchical fashion, explaining its success. Moreover, we are able to understand and describe the policies learned by DQNs for three different Atari2600 games and suggest ways to interpret, debug and optimize deep neural networks in reinforcement learning.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyzahavy2016graying

Related papers

• Visualizing Dynamics: From T-sne To Semi-mdps (2016)0.00
• A Theoretical Analysis Of Deep Q-learning (2019)0.00
• Deep Q-learning: Theoretical Insights From An Asymptotic Analysis (2020)10.35
• Convergent And Efficient Deep Q Network Algorithm (2021)0.00
• Universal Approximation Theorem Of Deep Q-networks (2025)0.00
• DQN With Model-based Exploration: Efficient Learning On Environments With Sparse Rewards (2019)0.00
• On The Convergence And Sample Complexity Analysis Of Deep Q-networks With \(\epsilon\)-greedy Exploration (2023)3.58
• Modular Multi-objective Deep Reinforcement Learning With Decision Values (2017)10.74