Adaptive Multi-Teacher Multi-level Knowledge Distillation

Abstract

Knowledge distillation~(KD) is an effective learning paradigm for improving the performance of lightweight student networks by utilizing additional supervision knowledge distilled from teacher networks. Most pioneering studies either learn from only a single teacher in their distillation learning methods, neglecting the potential that a student can learn from multiple teachers simultaneously, or simply treat each teacher to be equally important, unable to reveal the different importance of teachers for specific examples. To bridge this gap, we propose a novel adaptive multi-teacher multi-level knowledge distillation learning framework~(AMTML-KD), which consists two novel insights: (i) associating each teacher with a latent representation to adaptively learn instance-level teacher importance weights which are leveraged for acquiring integrated soft-targets~(high-level knowledge) and (ii) enabling the intermediate-level hints~(intermediate-level knowledge) to be gathered from multiple teachers by the proposed multi-group hint strategy. As such, a student model can learn multi-level knowledge from multiple teachers through AMTML-KD. Extensive results on publicly available datasets demonstrate the proposed learning framework ensures student to achieve improved performance than strong competitors.

Neurocomputing 2020.07

References in corpus (7)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Distilling the Knowledge in a Neural Network
• FitNets: Hints for Thin Deep Nets
• Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer
• Multilingual Neural Machine Translation with Knowledge Distillation
• Deep Model Compression: Distilling Knowledge from Noisy Teachers
• Knowledge Adaptation: Teaching to Adapt

Cited by in corpus (1)

• Zero-shot Adversarial Quantization