# Explainability in Human-Agent Systems

## TL;DR

This paper develops a comprehensive taxonomy of explainability in Human-Agent Systems, addressing fundamental questions about its purpose, audience, types, timing, and evaluation methods to advance understanding and implementation.

## Contribution

It introduces a detailed taxonomy of explainability, clarifies key concepts, and discusses evaluation strategies in Human-Agent Systems, filling gaps in conceptual understanding.

## Key findings

- Defines explainability and related terms clearly
- Provides a framework for when and how explanations should be provided
- Suggests methods for evaluating explainability systems

## Abstract

This paper presents a taxonomy of explainability in Human-Agent Systems. We consider fundamental questions about the Why, Who, What, When and How of explainability. First, we define explainability, and its relationship to the related terms of interpretability, transparency, explicitness, and faithfulness. These definitions allow us to answer why explainability is needed in the system, whom it is geared to and what explanations can be generated to meet this need. We then consider when the user should be presented with this information. Last, we consider how objective and subjective measures can be used to evaluate the entire system. This last question is the most encompassing as it will need to evaluate all other issues regarding explainability.