Why Self-Supervised Encoders Want to Be Normal

arXiv:2604.27743v1 Announce Type: cross
Abstract: We develop a geometric and information-theoretic framework for encoder-decoder learning built on the Information Bottleneck (IB) principle. Recasting IB as a rate-distortion problem with Kullback-Leibler (KL) divergence as distortion, we show that the optimal representation at any distortion level is a soft clustering of the emphpredictive manifold $mathcalM=p(Y$ inside the probability simplex, admitting a linear decoder in the canonical parameterization.
We derive a chain of exact transformations, from flat Dirichlet to exponential to isotropic Gaussian, connecting the maximum entropy prior on the simplex to Euclidean space, with quantified entropy overhead at each step, and show that Sketched Isotropic Gaussian Regularization (SIGReg) implements a Gaussian relaxation of this principle whose overhead affects rate accounting but not achievable prediction. This relaxation provides a principled distributional regularizer for learning with limited or no supervision.
Using the Conditional Entropy Bottleneck (CEB) decomposition, we derive concrete encoder losses for supervised and semi-supervised settings, estimated via minibatch marginals without variational bounds. In the self-supervised setting, the CEB conditional rate is replaced by a view-prediction proxy. SIGReg serves as the distributional regularizer for both the semi-supervised and self-supervised settings.
Experiments on toy problems and FashionMNIST confirm the predicted rate-distortion trade-offs and show that the non-parametric estimator is competitive with the standard variational approach.