search

bnu-wangxun / Deep_Metric

Deep Metric Learning
Apache License 2.0
778 stars 148 forks source link

About the Binomial Deviance Loss #30

Open Jeff-Zilence opened 5 years ago

Jeff-Zilence commented 5 years ago

Thank you for sharing your code. Nice work! I have a question about the Binomial Deviance Loss. In your previous paper, Binomial without mining achieves recall-top1 of 64% on CUB200, but I can not reproduce the result with your code. Could you provide more detail about your implementation? For example, the alpha and beta.

bnu-wangxun commented 5 years ago

alpha is 40 and beta is 2. Batchsize is 60-80, and adam with 1e-5 lr.

But performance on CUB is not stable (2% higher or lower is not suprising). The other three datasets is much easier to have the performance as in the paper.

Jeff-Zilence commented 5 years ago

Thank you very much. I reproduced the result of Binomial without mining. I think the problem is the normalization term (1/alpha and 1/beta). These terms are removed in your non-mining version and the accuracy starts dropping when I add these terms. But you use these normalization terms in the mining version. Is that designed for balancing the gradient of positive and negative pairs?

bnu-wangxun commented 5 years ago

Yes! Exactly that!