2 papers
cs.LG2026
OrthoReg: Orthogonal Regularization for Hybrid Symbolic-Neural Dynamical Systems
Till Richter, Niki Kilbertus
Dynamical systems are fundamental to modeling the natural world, yet modeling them involves a persistent trade-off: manually prescribed mechanistic models are interpretable by desi…
cs.LG2026
The Loss Is Not Enough: Sampling Conditions and Inductive Bias in Contrastive Representation Learning
Justinas Zaliaduonis, Patrick Putzky, Till Richter +1
Contrastive learning has become a leading paradigm for self-supervised representation learning, yet the conditions under which it recovers meaningful latent geometry remain incompl…