Comments
There's unfortunately not much to read here yet...
Knowledge Distillation (KD) aims to distill the knowledge of a cumbersome teacher model into a lightweight student model. Its success is generally attributed to the privileged information on similarities among categories provided by the teacher model, and in this sense, only strong teacher models are deployed to teach weaker students in practice. In this work, we challenge this common belief by following experimental observations: 1) beyond the acknowledgment that the teacher can improve the student, the student can also enhance the teacher significantly by reversing the KD procedure; 2) a poorly-trained teacher with much lower accuracy than the student can still improve the latter significantly. To explain these observations, we provide a theoretical analysis of the relationships between KD and label smoothing regularization.
There's unfortunately not much to read here yet...
Ever having issues keeping up with everything that's going on in Machine Learning? That's where we help. We're sending out a weekly digest, highlighting the Best of Machine Learning.
Discover the best guides, books, papers and news in Machine Learning, once per week.