Workshop Overview:
The rapid development and deployment of autonomous systems in the open world has highlighted the need for advances in the theory and practice of robust and reliable AI. Autonomous cars, drone delivery services, health care robots, and emergency response systems are some examples in which system decisions directly impact human lives. This workshop aims to bring together researchers from academia and industry to discuss the challenges involved in deployed autonomous systems, particularly systems that operate in the presence of minor perturbations or noise in the environment, and under model imprecision and uncertainty. We aim to highlight recent advances in theory and emerging application areas that require robust and reliable decision-making.
The workshop will offer a forum for researchers to discuss progress, identify knowledge gaps, and learn from case studies in robust and reliable decision-making, drawing on a wide range of methods, including automated planning, reinforcement learning, decision-theory, formal verification, multi-agent systems, game theory, robotics, AI ethics, and human factors. We welcome the participation of researchers from different disciplines representing different perspectives on the topic.
Relevant research challenges include (but are not limited to):
- Definitions of safety, robustness, reliability, and resilience
- Evaluation metrics for robustness and reliability, under model imprecision
- Decision-making representations, models, and algorithms for the open world
- Techniques to achieve resilient decision-making, under unmodelled disturbances
- Techniques to recognize and avoid negative side effects of AI systems
- Techniques for ethical, interpretable, fair, and trustworthy decision-making
- Case studies of robustness and reliability in deployed autonomous systems
- Learning to improve robustness and reliability from human feedback
- Models and algorithms to support competence-aware autonomy
Organizing Committee:
- Nick Hawes, University of Oxford
- Ece Kamar, Microsoft Research
- Bruno Lacerda (Co-chair), University of Oxford
- Sandhya Saisubramanian (Co-chair, primary contact), University of Massachusetts Amherst
- Shlomo Zilberstein, University of Massachusetts Amherst