Loss Weight ablation experiment

Question

Loss Weight ablation experiment

JinYu1998 opened this issue a year ago · 3 comments

JinYu1998 commented a year ago

Have you tried the effect of different loss weights on the distillation results ?

Answer 1 · 2024-02-05T11:56:30.000Z

Hey @JinYu1998 - we did a coarse sweep over the KL weights when setting up preliminary experiments on just the LibriSpeech corpus. We found the setting from DistilBART to be best, and so committed to this for the rest of the project. We didn't do any further tuning of the loss weights on our full training set. You can find an ablation over the loss terms (not weights) in page 26 of the paper.

Answer 2 · 2024-02-05T12:01:12.000Z

Hey @JinYu1998 - we did a coarse sweep over the KL weights when setting up preliminary experiments on just the LibriSpeech corpus. We found the setting from DistilBART to be best, and so committed to this for the rest of the project. We didn't do any further tuning of the loss weights on our full training set. You can find an ablation over the loss terms (not weights) in page 26 of the paper.

Thank you for your reply, I have previously worked on dynamic temperature distillation on classifieds, and just recently finished this work. I'm very interested in distillation in whisper, and look forward to combining my work with distill whisper very well.

Answer 3 · 2024-02-25T15:08:46.000Z

Delicious & Exciting Diet Foods : Weight Loss Food full video - https://youtu.be/4Kr8gtd2oss?si=1HVNwuBNTKgr4XCL