Understanding the Role of the Projector in Knowledge Distillation
This work addresses the problem of improving knowledge distillation efficiency and performance for researchers and practitioners in machine learning, though it is incremental as it builds on existing techniques.
The paper investigates the role of the projector in knowledge distillation, showing that it encodes past examples and influences training dynamics, leading to superior or comparable performance to state-of-the-art methods with higher efficiency, such as achieving 77.2% top-1 accuracy with DeiT-Ti on ImageNet.
In this paper we revisit the efficacy of knowledge distillation as a function matching and metric learning problem. In doing so we verify three important design decisions, namely the normalisation, soft maximum function, and projection layers as key ingredients. We theoretically show that the projector implicitly encodes information on past examples, enabling relational gradients for the student. We then show that the normalisation of representations is tightly coupled with the training dynamics of this projector, which can have a large impact on the students performance. Finally, we show that a simple soft maximum function can be used to address any significant capacity gap problems. Experimental results on various benchmark datasets demonstrate that using these insights can lead to superior or comparable performance to state-of-the-art knowledge distillation techniques, despite being much more computationally efficient. In particular, we obtain these results across image classification (CIFAR100 and ImageNet), object detection (COCO2017), and on more difficult distillation objectives, such as training data efficient transformers, whereby we attain a 77.2% top-1 accuracy with DeiT-Ti on ImageNet. Code and models are publicly available.