My goal is to design approaches that can learn, reason, and generalise from limited resources. My research focuses on Inductive Logic Programming (ILP), a form of program synthesis that uses logic to automatically learn verifiable logic programs from a small number of examples. I leverage constraint programming techniques to improve the learning performance of ILP. I co-develop the ILP system Popper.
I am leaving the University of Oxford and joining the University of Southampton. As a result, my Oxford email address is no longer active. You can contact me on the following address: celinehocquette@gmail.com.
Academic Employment
-
Lecturer, since 2024
University of Southampton -
Postdoc, 2021-2024
University of Oxford -
Ph.D., 2017-2021
Imperial College London -
11/03/25: Our paper An empirical comparison of cost functions in inductive logic programming, with Andrew Cropper, is available on arXiv.
-
20/02/25: I have been invited to present (virtually) at CEA LIST.
-
21/01/25: I presented (virtually) at INRIA, BOREAL.
-
2/12/24: I am joining the University of Southampton as a lecturer!
-
26-27/11/24: I have been invited to present at the Program Synthesis Days at CNRS LaBRI in Bordeaux.
-
21-22/11/24: I visited the Coconut group at the LIRMM laboratory (CNRS) in Montpellier, France.
-
28/10/24: I presented our poster Learning from raw data with low-level symbols alone, with Andrew Cropper, at the Beyond the symbols vs signals debate of the Royal Society, London.
-
22/10/24: I presented our paper Learning logic programs by finding minimal unsatisfiable subprograms at ECAI. Join me in the Reasoning and Learning 2 session to hear about this work!
-
07/08/24: I presented our papers Learning big logical rules by joining small rules and Learning logic programs by discovering higher-order abstractions at IJCAI. Join me in the KRR session to hear about these works!
-
08/07/24: I have been invited to present at the French National Research Institute for Agriculture, Food and the Environment (INRAE), on the topic An introduction of machine learning for livestock farming.
- 04/07/24: Our paper Learning logic programs by finding minimal unsatisfiable subprograms, with Andrew Cropper, has been accepted at ECAI.
-
24-27/06/24: I have been invited for a lecture at the summer school on symbolic and statistical methods for reasoning
and processing of formal expressions at Kutaisi International University.
-
13/06/24: My blog post Learning programs with numerical reasoning has been published at AIhub.org.
-
29/05/2024: I have been invited to present at the RuleML webinar.
-
01/05/2024: Our paper Can humans teach machines to code? with Johannes Langer, Andrew Cropper, and Ute Schmid, is available on arxiv.
-
16/04/2024: Our papers Learning big logical rules by joining small rules, with Andreas Niskanen, Rolf Morel, Matti Järvisalo, and Andrew Cropper, and Learning logic programs by discovering higher-order abstractions, with Sebastijan Dumančić and Andrew Cropper, have been accepted at IJCAI!