Jochen Triesch

Short info

My goal is to better understand how brains can learn so much more autonomously than current AI systems and how we can mimic this ability in future AIs. Taking inspiration from the cognitive development of infants and children, we are constructing computational models of human learning processes and let AIs grow up and learn in simulated physical environments. We see this „Developmental AI” as a promising route towards a more human-like general intelligence. Next to this basic research, we are also developing new machine learning techniques to help us better understand brain data, in particular in the context of brain disorders.

Open Science

Falck, J., Zhang, L., Raffington, L., Mohn, J. J., Triesch, J., Heim, C., & Shing, Y. L. (2023).
Longitudinal Changes in Value-based Learning in Middle Childhood: Distinct Contributions of Hippocampus and Striatum.
bioRxiv preprint

Xu, X., & Triesch, J. (2023).
CIPER: Combining Invariant and Equivariant Representations Using Contrastive and Predictive Learning.
arXiv preprint

Articles

Aubret, A., Ernst, M.R., Teulière, C., & Triesch, J. (2023).
Time to Augment Contrastive Learning.
IEEE Int. Conf. on Learning Representations (ICLR).

Aubret, A., Teulière, C., & Triesch, J. (2022).
Toddler-inspired embodied vision for learning object representations.
IEEE Int. Conf. on Development and Learning (ICDL).

Mattern, D., López, F.M., Ernst, M.R., Aubret, A., & Triesch, J. (2022).
MIMo: A Multi-Modal Infant Model for Studying Cognitive Development in Humans and AIs.
IEEE Int. Conf. on Development and Learning (ICDL), accepted.

Schneider, F., Xu, X., Ernst, M. R., Yu, Z., & Triesch, J. (2021).
Contrastive Learning Through Time.
In SVRHM 2021 Workshop@ NeurIPS.