PC-GZSL: Prior Correction for Generalized Zero Shot Learning

Generalized Zero Shot Learning (GZSL) aims at achieving a good accuracy on both seen and unseen classes by relying on the information acquired from auxiliary at-tributes. Existing approaches have devised many frame-works to make this knowledge transfer more efficient and in-formative. Despite their effectiveness on boosting the over-all performance, there has always been a strong bias in the model towards the seen classes which makes GZSL prob-lem more challenging. The effect of this bias on the model performance has never been properly explored. We observe that GZSL algorithms in literature have an evident bias to-wards the seen classes. Further we also show that techniques like calibrated stacking [7] fall short of resolving this conflict between the seen and unseen classes effectively. In this work we analyze and develop a logit-adjustment approach in GZSL setting and propose a simple, yet effective method to remove the bias from trained models in a post-hoc manner. Moreover, as a consequence of the post-hoc nature of the proposed approach, there is no additional training cost. We exhaustively compare the proposed method on both embedding-based and generative-based GZSL frameworks surpassing the SOTA results by 3.1%, 4.6% and 3.1% on CUB, SUN and AwA2 datasets. We also present theoretical analysis showing effectiveness ofproposed approach.