# Explainable Deep Reinforcement Learning Using Introspection in a Non-episodic Task

arXiv: https://arxiv.org/abs/2108.08911 · HTML: https://mlsanity.com/paper/2108.08911/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 10 references that are in the map.
- First posted on arXiv 18 Aug 2021.

## Builds on (most cited first, 5 of 5)

- [Rainbow: Combining Improvements in Deep Reinforcement Learning](https://mlsanity.com/paper/1710.02298/) (2017)
- [Dueling Network Architectures for Deep Reinforcement Learning](https://mlsanity.com/paper/1511.06581/) (2015)
- [Noisy Networks for Exploration](https://mlsanity.com/paper/1706.10295/) (2017)
- [Explainable robotic systems: Understanding goal-driven actions in a reinforcement learning scenario](https://mlsanity.com/paper/2006.13615/) (2020)
- [iCub: Learning Emotion Expressions using Human Reward](https://mlsanity.com/paper/2003.13483/) (2020)

## Built on by (most cited first, 2 of 2)

- [Introspection-based Explainable Reinforcement Learning in Episodic and Non-episodic Scenarios](https://mlsanity.com/paper/2211.12930/) (2022)
- [Explaining Agent's Decision-making in a Hierarchical Reinforcement Learning Scenario](https://mlsanity.com/paper/2212.06967/) (2022)

## Abstract (arXiv)

Explainable reinforcement learning allows artificial agents to explain their behavior in a human-like manner aiming at non-expert end-users. An efficient alternative of creating explanations is to use an introspection-based method that transforms Q-values into probabilities of success used as the base to explain the agent's decision-making process. This approach has been effectively used in episodic and discrete scenarios, however, to compute the probability of success in non-episodic and more complex environments has not been addressed yet. In this work, we adapt the introspection method to be used in a non-episodic task and try it in a continuous Atari game scenario solved with the Rainbow algorithm. Our initial results show that the probability of success can be computed directly from the Q-values for all possible actions.