Why the experimental results are inferior to those reported in paper?

[wangxu-scu]

Hi, I have checked the code, and I think everything is ok. But the results are inferior to those reported in the paper. For example, the accuracy for CIFAR-10 with this code is about 60% (DeepInfoMax-Local), while that in the paper is about 70%.
Do you have any idea that why there is such a big difference?

[DuaneNielsen]

Yeah, I think it's probably due to hyperparameter tuning. I did no tuning at all. For example, I used a very large batch size, perhaps a smaller batch size with more epochs would do better. Duane

…

On Tue, Oct 22, 2019 at 12:29 AM wangxu-scu ***@***.***> wrote: Hi, I have checked the code, and I think everything is ok. But the results are inferior to those reported in the paper. For example, the accuracy for CIFAR-10 with this code is about 60% (DeepInfoMax-Local), while that in the paper is about 70%. Do you have any idea that why there is such a big difference? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#7?email_source=notifications&email_token=ACUOYLAVI2CF4JY64MPGONLQP2TUNA5CNFSM4JDLVONKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HTNITSQ>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACUOYLHHG7YRUINEV2RQJXLQP2TUNANCNFSM4JDLVONA> .

-- Duane Broadcom 480 760 1559

[wangxu-scu]

thank you!