Simultaneously Learning Architectures and Features of Deep Neural Networks
This addresses the challenge of model compression for efficient deployment in various domains, though it appears incremental as it builds on existing pruning and regularization techniques.
The paper tackles the problem of compressing deep neural networks by simultaneously learning architectures and features, achieving improved trade-offs between model sizes and accuracies across applications like image classification, image compression, and audio classification.
This paper presents a novel method which simultaneously learns the number of filters and network features repeatedly over multiple epochs. We propose a novel pruning loss to explicitly enforces the optimizer to focus on promising candidate filters while suppressing contributions of less relevant ones. In the meanwhile, we further propose to enforce the diversities between filters and this diversity-based regularization term improves the trade-off between model sizes and accuracies. It turns out the interplay between architecture and feature optimizations improves the final compressed models, and the proposed method is compared favorably to existing methods, in terms of both models sizes and accuracies for a wide range of applications including image classification, image compression and audio classification.