Congratulations Martin! Outstanding doctoral thesis award
Martin was awarded an outstanding doctoral thesis award (top 5% of PhD graduate).
Martin first joined the team as an internship student from the French engineering school “Ecole des Ponts”. After 6 months discovering our research environment, Martin decided to embark onto a PhD.
During his PhD, Martin published 4 papers. I am particularly fond of two papers he led, because they exemplify the type of AI research needed to tackle challenging problems in applied ecology:
- Peron, M., K. H. Becker, P. Bartlett, and I. Chadès. 2017. Fast-tracking Stationary MOMDPs for Adaptive Management Problems. Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17). PDF
Martin developped a way of initialising M/POMDP solvers with better initial values for adaptive management problems which leads to improve performance. We are currently using Martin’s approach to solve large size adaptive management problems.
- Péron, M., P. L. Bartlett, K. H. Becker, K. Helmstedt, and I. Chadès. 2018. Two Approximate Dynamic Programming Algorithms for Managing Complete SIS Networks. Page 8. Proceedings of the 1st ACM SIGCAS Conference on Computing and Sustainable Societies. ACM. PDF
Susceptible-Infected-Susceptible (SIS) dynamic models are widely used to represent the way disease, invasive species or threatened species evolve over time (spread or go extinct locally). An issue arises when trying to prioritise allocation of resources across SIS networks: as the size and/or the connectivity of the network increase the problem becomes intractable. Martin studied the property of SIS networks and derived 2 algorithms that can solve completely connected SIS networks at a small loss of performance.
You can download Martin’s PhD here:
Péron, Martin Brice (2018) Optimal sequential decision-making under uncertainty. PhD by Publication, Queensland University of Technology. PDF