I’m interested in developing autonomous and embodied agents that can learn without supervision. This includes self-supervised representation learning, intrinsic motivation, hierarchical reinforcement learning and open-ended learning. I currently work as a postdoctoral researcher at the Frankfurt Institute for Advanced Studies (FIAS). Below is my list of papers:
Postdoc, Frankfurt Institute for Advanced Studies: Toddler-inspired object representation learning
Aubret, A., Teulière, C., & Triesch, J. (2024). Self-supervised visual learning from interactions with objects. The European Conference on Computer Vision, 2024. github
Aubret, A., Schaumlöffel, T., Roig, G., & Triesch, J. (2024). Learning Object Semantic Similarity with Self-Supervision. In 2024 IEEE International Conference on Development and Learning. IEEE. github to be coming
Ernst, M. R., López, F. M., Aubret, A., Fleming, R. W., & Triesch, J. (2024). Self-Supervised Learning of Color Constancy. In 2024 IEEE International Conference on Development and Learning. IEEE. github
Schaumlöffel, T., Aubret, A., Roig, G., & Triesch, J. (2023, November). Caregiver Talk Shapes Toddler Vision: A Computational Study of Dyadic Play. In 2023 IEEE International Conference on Development and Learning. IEEE. github to be coming
Postdoc, Clermont Ferrand Pascal institute: Toddler-inspired object representation learning
Aubret, A., Ernst, M. R., Teulière, C., & Triesch, J. Time to augment self-supervised visual representation learning. In The Eleventh International Conference on Learning Representations, 2023.
Arthur Aubret, Céline Teulière, and Jochen Triesch. Embodied vision for learning object representations. Joint IEEE International Conference on Development and Learning (ICDL), 2022.
Arthur Aubret, Céline Teulière, and Jochen Triesch. Toddler-inspired learning induces hierarchical object representations. 3th Sensorimotor Interaction, Language and Embodiment of Symbols (SMILES) workshop at ICDL, 2022.
Aubret, A., Lefort, M., Teulière, C., Matignon, L., Hassas, S., & Triesch, J. Compressed information is all you need: unifying intrinsic motivations and representation learning. In NeurIPS 2022 Workshop on Information-Theoretic Principles in Cognitive Systems, 2022.
Dominik Mattern, Francisco M. López, Markus R. Ernst, Arthur Aubret, and Jochen Triesch. Mimo: A multi-modal infant model for studying cognitive development in humans and ais. Joint IEEE International Conference on Development and Learning (ICDL), 2022.
Ph.D, Lyon, LIRIS: Learning increasingly complex skills through deep reinforcement learning using intrinsic motivation
Aubret, A., Matignon, L., & Hassas, S. (2023). An information-theoretic perspective on intrinsic motivation in reinforcement learning: a survey. Entropy, 2023.
Aubret, A., Matignon, L., Hassas, S.: DisTop: Discovering a Topological representation to learn diverse and rewarding skills. IEEE Transactions on Cognitive and Developmental Systems 2023.
Arthur Aubret, Laëtitia Matignon, Salima Hassas: ELSIM: End-to-End Learning of Reusable Skills Through Intrinsic Motivation. ECML/PKDD (2) 2020: 541-556
Aubret, A., Matignon, L., Hassas, S.: ELSIM: End-to-end learning of reusable skills through intrinsic motivation. ICML 2020 Workshop LifelongML
Aubret, A., Matignon, L., Hassas, S.: Étude de la motivation intrinsèque en apprentissage par renforcement. Journées Francophones Planification, Décision et Apprentissage, 2019.
Link thesis
Contact
Mail: ajp.aubret@gmail.com
Twitter: @Arthur_Aubret