On Difficulties of Probability Distillation
Probability distillation has recently been of interest to deep learning practitioners as it presents a practical solution for sampling from autoregressive models for deployment in real-time applications. We identify a pathological optimization issue with the commonly adopted stochastic minimization of the (reverse) KL divergence, owing to sparse gradient signal from the teacher model due to curse of dimensionality. We also explore alternative principles for distillation, and show that one can achieve qualitatively better results than with KL minimization.
PDF AbstractTasks
Datasets
MNIST
Fashion-MNIST
Results from the Paper
Submit
results from this paper
to get state-of-the-art GitHub badges and help the
community compare results to other papers.
Methods
No methods listed for this paper. Add
relevant methods here
