Skip to yearly menu bar Skip to main content


Efficient Testable Learning of Halfspaces with Adversarial Label Noise

Ilias Diakonikolas · Daniel Kane · Vasilis Kontonis · Sihan Liu · Nikos Zarifis

Great Hall & Hall B1+B2 (level 1) #1714
[ ]
Thu 14 Dec 8:45 a.m. PST — 10:45 a.m. PST

Abstract: We give the first polynomial-time algorithm for the testable learning of halfspaces in the presence of adversarial label noise under the Gaussian distribution. In the recently introduced testable learning model, one is required to produce a tester-learner such that if the data passes the tester, then one can trust the output of the robust learner on the data. Our tester-learner runs in time $\text{poly}(d/\epsilon)$ and outputs a halfspace with misclassification error $O(\text{opt})+\epsilon$, where $\text{opt}$ is the 0-1 error of the best fitting halfspace. At a technical level, our algorithm employs an iterative soft localization technique enhanced with appropriate testers to ensure that the data distribution is sufficiently similar to a Gaussian. Finally, our algorithm can be readily adapted to yield an efficient and testable active learner requiring only $d ~ \text{polylog}(1/\epsilon)$ labeled examples.

Chat is not available.