constraint_coefficient of surrogate loss for pulling selected keys closer

google-research / l2p

Learning to Prompt (L2P) for Continual Learning @ CVPR22 and DualPrompt: Complementary Prompting for Rehearsal-free Continual Learning @ ECCV22

https://arxiv.org/pdf/2112.08654.pdf

Apache License 2.0

416 stars 42 forks source link

Closed kimwongyuda closed 2 years ago

kimwongyuda commented 2 years ago

The constraint_coefficient of surrogate loss is 0.5 in the paper.

But, in the codes, that is set to -0.1.

What is correct? Which is better?

Thank you.

KingSpencer commented 2 years ago

Hi, thanks for pointing out this issue. I believe 0.5 is the correct value.