Entropic Autoencoders mitigate posterior collapse by implicitly defining priors via entropy in a free-energy-minimizing encoder ensemble, yielding multimodal latent distributions that preserve data structure on reaction-diffusion, MNIST, and CelebA.
Entropy-SGD: Biasing gradient descent into wide valleys*.Journal of Statistical Mechanics: Theory and Experiment, 2019 (12):124018, December 2019
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
For 2-layer homogeneous NNs on multi-index models, flattest interpolators achieve low population loss when data is a sum of single-index models with low approximation error and label noise.
citing papers explorer
-
Entropic Auto-Encoding via Implicit Free-Energy Minimization
Entropic Autoencoders mitigate posterior collapse by implicitly defining priors via entropy in a free-energy-minimizing encoder ensemble, yielding multimodal latent distributions that preserve data structure on reaction-diffusion, MNIST, and CelebA.
-
Flatness and Generalization: Learning Multi-Index Models with Homogeneous Neural Networks
For 2-layer homogeneous NNs on multi-index models, flattest interpolators achieve low population loss when data is a sum of single-index models with low approximation error and label noise.