disagreement notion
Beyond Top-Class Agreement: Using Divergences to Forecast Performance under Distribution Shift
Schirmer, Mona, Zhang, Dan, Nalisnick, Eric
Knowing if a model will generalize to data 'in the wild' is crucial for safe deployment. To this end, we study model disagreement notions that consider the full predictive distribution - specifically disagreement based on Hellinger distance, Jensen-Shannon and Kullback-Leibler divergence. We find that divergence-based scores provide better test error estimates and detection rates on out-of-distribution data compared to their top-1 counterparts. Experiments involve standard vision and foundation models.
2312.08033
Country:
- Europe > Netherlands > North Holland > Amsterdam (0.04)
- Europe > Portugal > Porto > Porto (0.04)
- Europe > Germany > Baden-Württemberg > Tübingen Region > Tübingen (0.04)