WepresentanewclassofGAMs thatusetensor rank decompositions of polynomials to learn powerful,inherently-interpretable models. Our approach, titled Scalable Polynomial Additive Models (SPAM) is effortlessly scalable and modelsall higher-order feature interactions without a combinatorial parameter explosion. SPAM outperforms allcurrent interpretable approaches, and matches DNN/XGBoost performance onaseries ofreal-world benchmarks with up to hundreds of thousands of features.

The probes are applied on datasets of real images with annotations for the property.

Many real-world vision problems suffer from inherent ambiguities. In clinical applications for example, itmight not be clear from aCT scan alone which particular region is cancer tissue. Therefore a group of graders typically produces a set of diverse but plausible segmentations.

Thus, fMRI BOLD activity reflects an indirect, noisy measure ofneural activity,mismatched toneural timescales byseveral orders ofmagnitude.

As a result, our method significantly accelerates ViTs for dense prediction tasks.

We give the first polynomial-time algorithm for robust regression in the listdecodable setting where an adversary can corrupt a greater than1/2 fraction ofexamples.

Neural Information Processing Systems http://nips.cc/