λ-JEPA: Spectral Anti-Collapse Regularization for Self-Supervised Learning
arXiv 2026
Berker Demirel
PhD candidate at the Institute of Science and Technology Austria, working on representation learning.
About
I am a PhD candidate at the Causal Learning and Artificial Intelligence Lab at IST Austria, supervised by Francesco Locatello. My research is on representation learning, with a focus on how neural networks organize information, when their representations collapse and how they behave under distribution shift. I am also interested in self-supervised learning, generalization and generative modeling.
I am co-affiliated with the Chan Zuckerberg Initiative's AI for Science program, which supports my PhD research.
Prior to beginning my PhD, I earned a master's degree from Sabanci University. During this time, I was a member of the VPA Lab and was under the supervision of Asst. Prof. Huseyin Ozkan and Asst. Prof. Erchan Aptoula. The primary focus of my master's studies was on group activity recognition and domain generalization.
In the summer of 2019, I completed a summer internship at Erasmus University Rotterdam.