Research

Thinking about things.

I’m interested in developing safe and robust machine learning systems that behave in the way we expect, making them able to work with humans effectively.

One aspect of this is understanding decision making agents through imitation and inverse reinforcement learning. A probabilistic perspective allows us to handle the crucial uncertainty surrounding behaviour and as part of my PhD I’m looking to build scalable and well-calibrated Bayesian approximate methods that can bring us key insights and support high-stakes decisions particularly in the medical setting.

I’m also interested in general problems in reward (mis-)specification / hacking / (mis-)generalisation, as well as recently understanding internal concepts in large foundation models.

A full list of my published work can be found on my Google Scholar

Highlighted Publications

A. J. Chan & M. van der Schaar Synthetic Model Combination: An Instance-wise Approach to Unsupervised Ensemble Learning, Advances in Neural Information Processing Systems (NeurIPS), 2022.

A. J. Chan, A. Curth & M. van der Schaar Inverse Online Learning: Understanding Non-Stationary and Reactionary Policies, International Conference on Learning Representations (ICLR), 2022.

A. Pace, A. J. Chan, & M. van der Schaar POETREE: Interpretable Policy Learning with Adaptive Decision Trees, International Conference on Learning Representations (ICLR), 2022.

A. J. Chan, I. Bica, A. Huyuk, D. Jarrett, & M. van der Schaar The Medkit-learn(ing) Environment: Medical Decision Modelling through Simulation, Proceedings of the Neural Information Processing Systems (NeurIPS) track on Datasets and Benchmarks, 2021.

A. J. Chan & M. van der Schaar Scalable Bayesian Inverse Reinforcement Learning, International Conference on Learning Representations (ICLR), 2021.

A. M. Alaa, A. J. Chan, & M. van der Schaar Generative Time-series Modeling with Fourier Flows, International Conference on Learning Representations (ICLR), 2021.

A. J. Chan, A. M. Alaa, Z. Qian, & M. van der Schaar Unlabelled Data Improves Bayesian Uncertainty Calibration under Covariate Shift, International Conference on Machine Learning (ICML), 2020.

A. J. Chan & M. van der Schaar Interpretable Policy Learning, MPhil Machine Learning and Machine Intelligence Thesis, 2020.

A. J. Chan & R. Silva Probabilistic Deep Learning, BSc Statistics Thesis, 2019.