Breaking Inter-Layer Co-Adaptation by Classifier Anonymization
This addresses a specific issue in neural network optimization for researchers, but appears incremental as it builds on existing co-adaptation concerns.
The paper tackles the problem of co-adaptation between feature extractors and classifiers in neural networks, which can degrade test performance, and introduces FOCA, a method that uses randomly-generated weak classifiers to avoid this issue, with experiments showing supportive evidence.
This study addresses an issue of co-adaptation between a feature extractor and a classifier in a neural network. A naive joint optimization of a feature extractor and a classifier often brings situations in which an excessively complex feature distribution adapted to a very specific classifier degrades the test performance. We introduce a method called Feature-extractor Optimization through Classifier Anonymization (FOCA), which is designed to avoid an explicit co-adaptation between a feature extractor and a particular classifier by using many randomly-generated, weak classifiers during optimization. We put forth a mathematical proposition that states the FOCA features form a point-like distribution within the same class in a class-separable fashion under special conditions. Real-data experiments under more general conditions provide supportive evidences.