Critic complexity quantified by spectral effective-rank entropy is measurable during TD3/PPO training, associated with behavior in a heterogeneous way, and can be altered by adding a spectral-entropy penalty to the critic loss.
Complexity control facilitates reasoning-based compositional generalization in transform- ers
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2verdicts
UNVERDICTED 2representative citing papers
Neural networks exhibit condensation of neurons into clusters with similar outputs whose number increases monotonically during training, facilitated by small initializations or dropout, providing insights into generalization and reasoning.
citing papers explorer
-
Gauging, Measuring, and Controlling Critic Complexity in Actor-Critic Reinforcement Learning
Critic complexity quantified by spectral effective-rank entropy is measurable during TD3/PPO training, associated with behavior in a heterogeneous way, and can be altered by adding a spectral-entropy penalty to the critic loss.
-
An overview of condensation phenomenon in deep learning
Neural networks exhibit condensation of neurons into clusters with similar outputs whose number increases monotonically during training, facilitated by small initializations or dropout, providing insights into generalization and reasoning.