Improving Unsupervised Task-driven Models of Ventral Visual Stream via Relative Position Predictivity
This work addresses a gap in computational neuroscience by modeling additional functions of VVS beyond object recognition, though it is incremental as it builds on existing contrastive learning methods.
The paper tackles the limitation of current unsupervised task-driven models of the ventral visual stream (VVS) by integrating relative position (RP) prediction with contrastive learning, resulting in improved object recognition performance and enhanced brain similarity.
Based on the concept that ventral visual stream (VVS) mainly functions for object recognition, current unsupervised task-driven methods model VVS by contrastive learning, and have achieved good brain similarity. However, we believe functions of VVS extend beyond just object recognition. In this paper, we introduce an additional function involving VVS, named relative position (RP) prediction. We first theoretically explain contrastive learning may be unable to yield the model capability of RP prediction. Motivated by this, we subsequently integrate RP learning with contrastive learning, and propose a new unsupervised task-driven method to model VVS, which is more inline with biological reality. We conduct extensive experiments, demonstrating that: (i) our method significantly improves downstream performance of object recognition while enhancing RP predictivity; (ii) RP predictivity generally improves the model brain similarity. Our results provide strong evidence for the involvement of VVS in location perception (especially RP prediction) from a computational perspective.