AI Apology: An Apologetic Approach to Socially Responsible Agents

Hadassah Harland, 2nd-year CSIRO R+ top-up scholarship PhD student with Deakin University (supervised by Prof Richard Dazeley, Prof Peter Vamplew, Dr Bahareh Nakisa, Dr Francisco Cruz and Dr Hashini Senaratne), is studying how to design an apologetic framework for AI/robot agents. This project focuses on designing, implementing, and evaluating an impactful tool for machine agents to communicate awareness of any misaligned behaviors or harm caused during human-machine collaborative tasks and to repair trust. The design and implementation of this AI apology component will be based on hierarchical multi-objective reinforcement learning.