Dillon Ze Chen

        E: dillon dot chen 1 at gmail dot com
        T: +33 05 61 33 63 48
        W: https://dillonzchen.github.io

I'm currently a PhD student under the supervision of Sylvie Thiébaux and Arthur Bit-Monnot.

I was previously an undergraduate student at the Australian National University. I received the University Medal for my honours year in computer science, and the Hanna Neumann Prize for Mathematics for achieving the highest coursework marks in my cohort. I was fortunate enough to work with Pascal Bercher, Qing Wang, Felipe Trevizan and Sylvie Thiébaux.

For research, keep reading below. To play the dog gacha, refresh the page (rates and rarities coming soon!). My current bench press PB is 130kg.

Links

[dblp] [Google Scholar] [GitHub]

News

[2024.04]	Excited to visit CVUT in Prague. Thanks Rostislav for hosting!
[2024.02]	I am visiting AAAI 2024 in Vancouver.
[2024.02]	I am giving a talk at the ANU AI ML Friends seminar.
[2023.12]	I am visiting NeurIPS 2023 in New Orleans.
[2023.02]	I am visiting AAAI 2023 in Washington D.C.

Research

I am interested in the intersection of the fields of planning, symbolic AI and learning.

Planning is focused on long horizon reasoning and complex problem solving with guarantees. However, planning models suffer from computationally complexity. The tradeoff with planning is between expressivity and scalability: we want planning models which are expressive enough to model practical, real life problems, but are also solvable with reasonable computational resources to be actually useful.

Learning is focused on pattern matching and data compression. However, learning models generally do not have guarantees with their outputs. The tradeoff with learning is between expressivity and generalisability: we want learning models which are expressive enough to capture certain patterns, but are also able to generalise to unseen outputs.

I'm currently focused on building intelligent systems that can efficiently and effectively learn to compute plans that require complex reasoning.

Publications

Conference papers

[C7]	Dillon Ze Chen and Sylvie Thiébaux. Novelty Heuristics, Multi-Queue Search, and Portfolios for Numeric Planning. In 17th International Symposium on Combinatorial Search (SoCS), 2024. [ pdf \| poster \| slides \| code ]
[C6]	Dillon Ze Chen, Felipe Trevizan, and Sylvie Thiébaux. Return to Tradition: Learning Reliable Heuristics with Classical Machine Learning. In 34th International Conference on Automated Planning and Scheduling (ICAPS), 2024. [ pdf \| poster \| slides \| code ]
[C5]	Dillon Ze Chen, Sylvie Thiébaux, and Felipe Trevizan. Learning Domain-Independent Heuristics for Grounded and Lifted Planning. In 38th AAAI Conference on Artificial Intelligence, 2024. Oral. [ pdf \| poster \| slides \| code ]
[C4]	Qing Wang, Dillon Ze Chen, Asiri Wijesinghe, Shouheng Li, and Muhammad Farhan. N-WL: A New Hierarchy of Expressivity for Graph Neural Networks. In 11th International Conference on Learning Representations (ICLR), 2023. [ pdf \| poster \| slides \| code ]
[C3]	Dillon Ze Chen, Felipe Trevizan, and Sylvie Thiébaux. Heuristic Search for Multi-Objective Probabilistic Planning. In 37th AAAI Conference on Artificial Intelligence, 2023. Oral. [ pdf \| poster \| slides \| code ]
[C2]	Dillon Ze Chen and Pascal Bercher. The Complexity of Flexible FOND HTN Planning. In 32nd International Conference on Automated Planning and Scheduling (ICAPS), 2022. Best Undergraduate Student Paper Award. [ pdf \| poster \| slides ]
[C1]	Dillon Ze Chen and Pascal Bercher. Fully Observable Nondeterministic HTN Planning – Formalisation and Complexity Results. In 31st International Conference on Automated Planning and Scheduling (ICAPS), 2021. Best Undergraduate Student Paper Award. [ pdf \| poster \| slides ]

Workshop papers

[W3]	Dillon Ze Chen, Sylvie Thiébaux, and Felipe Trevizan. GOOSE: Learning Domain-Independent Heuristics. In 7th Workshop on Generalization in Planning (GenPlan), 2023. «Subsumed by [C5]». [ pdf \| poster \| slides ]
[W2]	Dillon Ze Chen, Felipe Trevizan, and Sylvie Thiébaux. Graph Neural Networks and Graph Kernels For Learning Heuristics: Is there a difference?. In 7th Workshop on Generalization in Planning (GenPlan), 2023. «Subsumed by [C6]». [ pdf \| poster ]
[W1]	Dillon Ze Chen and Pascal Bercher. The Complexity of Flexible FOND HTN Planning. In 4th ICAPS Workshop on Hierarchical Planning (HPlan), 2021. «Subsumed by [C2]». [ pdf ]

Clocks

Time zone map