RLMViz: Interpreting Deep Reinforcement Learning Memory - Archive ouverte HAL Access content directly
Conference Papers Year : 2019

RLMViz: Interpreting Deep Reinforcement Learning Memory

RLMViz: Interpréter la Mémoire du Deep Reinforcement Learning


We present RLMViz, a visual analytics interface to interpret the internal memory of an agent (e.g., a robot) trained using deep reinforcement learning. This memory is composed of large temporal vectors updated before each action of the agent moving in an environment. This memory is not trivial to understand, and is referred to as a black box, which only inputs (images) and outputs (actions) are understood, but not its inner workings. Using RLMViz, experts can form hypothesis on this memory and derive rules based on the agent's decisions to interpret them, and gain an understanding towards why errors have been made and improve future training process. We report on the main features of RLMViz which are memory navigation and contextualization techniques using time-lines juxtapositions. We also present our early findings using the VizDoom simulator, a standard benchmark for DRL navigation scenarios.
Fichier principal
Vignette du fichier
Journee_Visu_2019__DRLViz.pdf (1.03 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02140902 , version 1 (27-05-2019)


  • HAL Id : hal-02140902 , version 1


Theo Jaunet, Romain Vuillemot, Christian Wolf. RLMViz: Interpreting Deep Reinforcement Learning Memory. Journée Visu 2019, May 2019, Paris, France. ⟨hal-02140902⟩
170 View
143 Download


Gmail Facebook X LinkedIn More