Explainable AI
People are increasingly subject to decisions made wholly or in part by AI systems. But how can we trust that the decisions have been made fairly and reasonably?
It’s not enough for the computer to say no - opaque decision-making undermines trust, accountability, and effective oversight, and sits at odds with emerging AI regulation and established rights to an explanation for automated decisions.
ARAAC researchers work on making the behaviour and decisions of AI systems interpretable to the humans who use them, from agents that can explain the reasoning and trade-offs behind their choices, to representations and learning methods that are transparent by design. This work is closely tied to our research on safety and human-agent interaction, where understanding an agent’s intent is the foundation of trust.
Key Researchers
Dr Francisco Cruz
UNSW
Francisco’s current research interests include reinforcement learning, explainable artificial intelligence, human-robot interaction, artificial neural networks, and psychologically and bio-inspired models.
Professor Richard Dazeley
Deakin University
Richard is the Leader of the Machine Intelligence Lab at Deakin University (Geelong), and the Deputy Head of School. He is a leading researcher in the Human-alignment of autonomous agents through Safe, Ethical, Explainable and Interactive methods utilising Multiobjective Reinforcement Learning (MORL) and is a senior member of the AI existential Safety Community
Associate Professor Sunil Aryal
Deakin University
Sunil works in a wide range of areas in ML/AI from similarity measures, anomaly detection, classification, clustering, text/image analysis, ensemble learning, learning from small subsamples of data, and robust/explainable machine learning to autonomous systems.
ARAAC Publications
- (2025). Understanding user preferences in explainable artificial intelligence: a mapping function proposal ACM Transactions on Intelligent Systems and Technology
- (2024). EMOTE: An Explainable Architecture for Modelling the Other through Empathy International Joint Conference on Artificial Intelligence (IJCAI)
- (2023). Explainable reinforcement learning for broad-XAI: a conceptual framework and survey Neural Computing and Applications
- (2022). Analysis of explainable goal-driven reinforcement learning in a continuous simulated environment Algorithms
- (2022). Evaluating human-like explanations for robot actions in reinforcement learning scenarios IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
- (2021). Explainable robotic systems: understanding goal-driven actions in a reinforcement learning scenario Neural Computing and Applications
- (2021). Levels of explainable artificial intelligence for human-aligned conversational explanations Artificial Intelligence
- (2019). Memory-based explainable reinforcement learning Australasian Joint Conference on Artificial Intelligence (AI 2019)