A hidden Markov model fit to aligned per-window classifier weights recovers temporal states, and models transfer better within those states than across boundaries.
Exact Phase Transitions in Deep Learning
1 Pith paper cite this work, alongside 1 external citations. Polarity classification is still indexing.
1
Pith paper citing it
1
external citations · Pith
abstract
This work reports deep-learning-unique first-order and second-order phase transitions, whose phenomenology closely follows that in statistical physics. In particular, we prove that the competition between prediction error and model complexity in the training loss leads to the second-order phase transition for nets with one hidden layer and the first-order phase transition for nets with more than one hidden layer. The proposed theory is directly relevant to the optimization of neural networks and points to an origin of the posterior collapse problem in Bayesian deep learning.
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Latent States in Neural Networks: Recovering the Temporal Structure of Drifting Data from Model Weights
A hidden Markov model fit to aligned per-window classifier weights recovers temporal states, and models transfer better within those states than across boundaries.