Timezone: »

Separation Results between Fixed-Kernel and Feature-Learning Probability Metrics
Carles Domingo i Enrich · Youssef Mroueh

Tue Dec 07 08:30 AM -- 10:00 AM (PST) @ None #None
Several works in implicit and explicit generative modeling empirically observed that feature-learning discriminators outperform fixed-kernel discriminators in terms of the sample quality of the models. We provide separation results between probability metrics with fixed-kernel and feature-learning discriminators using the function classes $\mathcal{F}_2$ and $\mathcal{F}_1$ respectively, which were developed to study overparametrized two-layer neural networks. In particular, we construct pairs of distributions over hyper-spheres that can not be discriminated by fixed kernel $(\mathcal{F}_2)$ integral probability metric (IPM) and Stein discrepancy (SD) in high dimensions, but that can be discriminated by their feature learning ($\mathcal{F}_1$) counterparts. To further study the separation we provide links between the $\mathcal{F}_1$ and $\mathcal{F}_2$ IPMs with sliced Wasserstein distances. Our work suggests that fixed-kernel discriminators perform worse than their feature learning counterparts because their corresponding metrics are weaker.

Author Information

Carles Domingo i Enrich (New York University)
Youssef Mroueh (IBM T.J Watson Research Center)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors