Felipe Trevizan

Dr. Felipe Trevizan is a Senior Lecturer in the School of Computing at the Australian National University (ANU), specializing in Artificial Intelligence, Operations Research, and Machine Learning. His research focuses on automated planning, scheduling, and heuristic search, particularly in uncertain environments. Previously, he was a Senior Research Scientist at NICTA/Data61 and earned his Ph.D. in Machine Learning from Carnegie Mellon University under Prof. Manuela Veloso.

Education:

• Ph.D. in Machine Learning, Carnegie Mellon University (2013)
• M.Sc. in Machine Learning, Carnegie Mellon University (2010)
• M.Sc. in Computer Science, University of São Paulo (2006)
• B.Sc. in Computer Science, University of São Paulo (2004)

Research Interests: Felipe’s work bridges AI and Operations Research, emphasizing automated planning under uncertainty, heuristic search algorithms, and integration of machine learning with planning systems. His contributions include novel methods like short-sighted planning and occupation measure heuristics, as well as advancements in multi-objective stochastic planning and constraint generation techniques.

Awards:

• 2016 Kikuchi-Karlaftis Best Paper Award (Transport Research Board)
• Best Paper Awards at ICAPS 2016 and 2017

Advising & Grants: Felipe has mentored over 20 students, including Ph.D. candidates Mingyu Hao and Johannes Schmalz, and Honours students Ryan Wang and Dillon Chen. His research is supported by grants focusing on planning under uncertainty, optimization in transportation, and machine learning integration with classical planning.

Key Contributions: Recent work includes partial-space search for learned heuristics, CG-iLAO* for stochastic shortest paths, and GOOSE, a domain-independent heuristic learning framework. These innovations aim to enhance planning efficiency and adaptability in complex domains.