# A Bayesian Approach to Identifying Representational Errors

arXiv: https://arxiv.org/abs/2103.15171 · HTML: https://mlsanity.com/paper/2103.15171/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 51 references that are in the map.
- First posted on arXiv 28 Mar 2021.

## Builds on (most cited first, 20 of 20)

- [Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks](https://mlsanity.com/paper/1511.06434/) (2015)
- [Reinforcement Learning with Unsupervised Auxiliary Tasks](https://mlsanity.com/paper/1611.05397/) (2016)
- [Cooperative Inverse Reinforcement Learning](https://mlsanity.com/paper/1606.03137/) (2016)
- [A Survey of Inverse Reinforcement Learning: Challenges, Methods and Progress](https://mlsanity.com/paper/1806.06877/) (2018)
- [On Deep Multi-View Representation Learning: Objectives and Optimization](https://mlsanity.com/paper/1602.01024/) (2016)
- [Near-Optimal Representation Learning for Hierarchical Reinforcement Learning](https://mlsanity.com/paper/1810.01257/) (2018)
- [Third-Person Imitation Learning](https://mlsanity.com/paper/1703.01703/) (2017)
- [SDRL: Interpretable and Data-efficient Deep Reinforcement Learning Leveraging Symbolic Planning](https://mlsanity.com/paper/1811.00090/) (2018)
- [Learning human behaviors from motion capture by adversarial imitation](https://mlsanity.com/paper/1707.02201/) (2017)
- [Latent Space Policies for Hierarchical Reinforcement Learning](https://mlsanity.com/paper/1804.02808/) (2018)
- [Modeling Others using Oneself in Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/1802.09640/) (2018)
- [Deep Object-Centric Representations for Generalizable Robot Learning](https://mlsanity.com/paper/1708.04225/) (2017)
- [Online Planning Algorithms for POMDPs](https://mlsanity.com/paper/1401.3436/) (2014)
- [Shared Autonomy via Hindsight Optimization for Teleoperation and Teaming](https://mlsanity.com/paper/1706.00155/) (2017)
- [Coordinated Exploration via Intrinsic Rewards for Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/1905.12127/) (2019)
- [PEGASUS: A Policy Search Method for Large MDPs and POMDPs](https://mlsanity.com/paper/1301.3878/) (2013)
- [The Value of Inferring the Internal State of Traffic Participants for Autonomous Freeway Driving](https://mlsanity.com/paper/1702.00858/) (2017)
- [Universal Successor Representations for Transfer Reinforcement Learning](https://mlsanity.com/paper/1804.03758/) (2018)
- [Object-sensitive Deep Reinforcement Learning](https://mlsanity.com/paper/1809.06064/) (2018)
- [Discovering Blind Spots in Reinforcement Learning](https://mlsanity.com/paper/1805.08966/) (2018)

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

- [Computing Policies That Account For The Effects Of Human Agent Uncertainty During Execution In Markov Decision Processes](https://mlsanity.com/paper/2109.07436/) (2021)
- [Aligning Robot Representations with Humans](https://mlsanity.com/paper/2205.07882/) (2022)

## Abstract (arXiv)

Trained AI systems and expert decision makers can make errors that are often difficult to identify and understand. Determining the root cause for these errors can improve future decisions. This work presents Generative Error Model (GEM), a generative model for inferring representational errors based on observations of an actor's behavior (either simulated agent, robot, or human). The model considers two sources of error: those that occur due to representational limitations -- "blind spots" -- and non-representational errors, such as those caused by noise in execution or systematic errors present in the actor's policy. Disambiguating these two error types allows for targeted refinement of the actor's policy (i.e., representational errors require perceptual augmentation, while other errors can be reduced through methods such as improved training or attention support). We present a Bayesian inference algorithm for GEM and evaluate its utility in recovering representational errors on multiple domains. Results show that our approach can recover blind spots of both reinforcement learning agents as well as human users.