Felipe Trevizan

Felipe W. Trevizan

Building 145, office 3.20
Australian National University Campus
Canberra ACT 2601, Australia

I am an assistant professor in the Artificial Intelligence Group at the Australian National University. My research interests lie at the intersection of Artificial Intelligence, Operations Research and Machine Learning including automated planning and scheduling, planning under uncertainty, heuristic search, and reinforcement learning.

Selected Publications (full list here)

  • Learning Planning Heuristics with Hypergraph Networks. (2020) Shen, W., Trevizan, F. and Thiébaux, S. In Proc. of 30th Int. Conf. on Automated Planning and Scheduling (ICAPS). [pdf] [bib]
  • ASNets: Deep Learning for Generalised Planning. (2020) Toyer, S., Thiébaux, S., Trevizan, F. and Xie, L. In Journal of Artificial Intelligence. [pdf] [bib]
  • Occupation Measure Heuristics for Probabilistic Planning. (2017) Trevizan, F., Thiébaux, S. and Haslum, P. In Proc. of 27th Int. Conf. on Automated Planning and Scheduling (ICAPS). (Best Paper Award) [pdf] [bib]
  • Heuristic Search in Dual Space for Constrained Stochastic Shortest Path Problems. (2016) Trevizan, F., Thiébaux, S., Santana, P. and Williams, B. In Proc. of 26th Int. Conf. on Automated Planning and Scheduling (ICAPS). (Best Paper Award) [pdf] [bib]
  • A Non-homogeneous Time Mixed Integer LP Formulation for Traffic Signal Control. (2016) Guilliard, I., Sanner, S., Trevizan, F. and Williams, B. In Transport Research Record (TRR): Journal of the Transport Research Board. (Kikuchi-Karlaftis Best Paper Award) [pdf] [bib] [demo]


  • Artificial Intelligence (COMP3620/6320): 2020, 2019
  • News and Highlights
    • STRIPS-HGN: First algorithm capable of learning domain-independent heuristics from scratch (ICAPS'20).
    • New insights and results for ASNets on JAIR
    • Guiding search using generalized policies from ASNets (SoCS'19)
    • PLTL-dual: First heuristic search algorithm for SSPs and MDPs with Probabilistic LTL constraints (KR'18).
    • ASNets: Learning generalized policies for SSPs using neural nets (AAAI'18).
    • h-pom, h-roc, and i2-dual won the best paper award at ICAPS'17!
    • New efficient approach to solve SSPs and C-SSPs with dead ends (UAI'17).
    • h-pom and h-roc: the first heuristics able to handle probabilities and costs for SSPs and C-SSPs (ICAPS'17).
    • i-dual won the best paper award at ICAPS'16!
    • QTM won the Kikuchi-Karlaftis best paper award at TRB'16! (demo).
  • Word cloud of my papers
  • wordcloud
    large wordcloud