GenEFT: Understanding Statics and Dynamics of Model Generalization via Effective Theory
This provides a physics-inspired approach to bridge theoretical predictions with practice in machine learning, though it appears incremental as it applies an existing effective theory concept to neural networks.
The paper tackles understanding neural network generalization by developing GenEFT, an effective theory framework that models latent-space representations as interacting particles, and finds it explains experimentally observed phase transitions between generalization and overfitting in graph learning examples.
We present GenEFT: an effective theory framework for shedding light on the statics and dynamics of neural network generalization, and illustrate it with graph learning examples. We first investigate the generalization phase transition as data size increases, comparing experimental results with information-theory-based approximations. We find generalization in a Goldilocks zone where the decoder is neither too weak nor too powerful. We then introduce an effective theory for the dynamics of representation learning, where latent-space representations are modeled as interacting particles (repons), and find that it explains our experimentally observed phase transition between generalization and overfitting as encoder and decoder learning rates are scanned. This highlights the power of physics-inspired effective theories for bridging the gap between theoretical predictions and practice in machine learning.