search

changlin31 / BossNAS

(ICCV 2021) BossNAS: Exploring Hybrid CNN-transformers with Block-wisely Self-supervised Neural Architecture Search
137 stars 20 forks source link

About formulation (1) and (6) #5

Closed NickChang97 closed 3 years ago

NickChang97 commented 3 years ago

Hi, very thanks for sharing your nice work. In the paper's formulation (1) and (6)， all has λ_k. But it seems to be no explaination about them. Could you please point it out here.

changlin31 commented 3 years ago

λ_k represent the weighting factor (a hyperparameter) to balance the loss of different blocks. These factors are set to 1 in this work and our previous work [36] and is learnable in DONNA [45].

NickChang97 commented 3 years ago

Really thanks for your reply.