Levels of explainable artificial intelligence for human-aligned conversational explanations

Abstract

Over the last few years there has been rapid research growth into eXplainable Artificial Intelligence (XAI) and the closely aligned Interpretable Machine Learning (IML). Drivers for this growth include recent legislative changes and increased investments by industry and governments, along with increased concern from the general public. People are affected by autonomous decisions every day and the public need to understand the decision-making process to accept the outcomes. However, the vast majority of the applications of XAI/IML are focused on providing low-level `narrow' explanations of how an individual decision was reached based on a particular datum. While important, these explanations rarely provide insights into an agent's: beliefs and motivations; hypotheses of other (human, animal or AI) agents' intentions; interpretation of external cultural expectations; or, processes used to generate its own explanation. Yet all of these factors, we propose, are essential to providing the explanatory depth that people require to accept and trust the AI's decision-making. This paper aims to define levels of explanation and describe how they can be integrated to create a human-aligned conversational explanation system. In so doing, this paper will survey current approaches and discuss the integration of different technologies to achieve these levels with Broad eXplainable Artificial Intelligence (Broad-XAI), and thereby move towards high-level `strong' explanations.

arXiv comment35 pages, 13 figures

References in corpus (31)

• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Towards A Rigorous Science of Interpretable Machine Learning
• Axiomatic Attribution for Deep Networks
• European Union regulations on algorithmic decision-making and a "right to explanation"
• Deep Facial Expression Recognition: A Survey
• Benchmarking Deep Reinforcement Learning for Continuous Control
• Model-Agnostic Interpretability of Machine Learning
• Towards Explainable Artificial Intelligence
• Deep Reinforcement Learning: An Overview
• Visualizing Deep Neural Network Decisions: Prediction Difference Analysis
• Recent Trends in Deep Learning Based Personality Detection
• Explanation in Human-AI Systems: A Literature Meta-Review, Synopsis of Key Ideas and Publications, and Bibliography for Explainable AI
• Emotion in Reinforcement Learning Agents and Robots: A Survey
• A Multi-Objective Deep Reinforcement Learning Framework
• Human Action Recognition and Prediction: A Survey
• Advances in Human Action Recognition: A Survey
• Multi-Objective Deep Reinforcement Learning
• Affect Control Processes: Intelligent Affective Interaction using a Partially Observable Markov Decision Process
• Machine Theory of Mind
• Dynamic Weights in Multi-Objective Deep Reinforcement Learning
• Attentive Explanations: Justifying Decisions and Pointing to the Evidence
• Trends in Integration of Vision and Language Research: A Survey of Tasks, Datasets, and Methods
• Automated Rationale Generation: A Technique for Explainable AI and its Effects on Human Perceptions
• Towards Transparent AI Systems: Interpreting Visual Question Answering Models
• Survey on Vision-based Path Prediction
• Interpretable & Explorable Approximations of Black Box Models
• Pedestrian Attribute Recognition: A Survey
• Blaming humans in autonomous vehicle accidents: Shared responsibility across levels of automation
• A Short Survey On Memory Based Reinforcement Learning
• Towards Interpretable R-CNN by Unfolding Latent Structures
• Joint Event Detection and Description in Continuous Video Streams

Cited by in corpus (7)

• Explainable, Domain-Adaptive, and Federated Artificial Intelligence in Medicine
• Directive Explanations for Monitoring the Risk of Diabetes Onset: Introducing Directive Data-Centric Explanations and Combinations to Support What-If Explorations
• Explainable robotic systems: Understanding goal-driven actions in a reinforcement learning scenario
• A Practical Guide to Multi-Objective Reinforcement Learning and Planning
• A three-Level Framework for LLM-Enhanced eXplainable AI: From technical explanations to natural language
• A Semi-Automated Solution Approach Recommender for a Given Use Case: a Case Study for AI/ML in Oncology via Scopus and OpenAI
• Experiential Explanations for Reinforcement Learning