… We introduce the notion of an (L, C)-omnipredictor, which could … In a sense, omnipredictors
extract all the predictive power from … an (L, C)-omnipredictor, where L contains all convex loss …

… Prior work studies omniprediction in the batch setting. We initiate the study of omniprediction
in … In contrast, omniprediction is most interesting for learning theoretic hypothesis classes F, …

… Omnipredictors are simple … for omniprediction, in both the online and offline settings.
To begin, we give an oracle-efficient online learning algorithm that acheives (L, H)omniprediction …

… on a simulated distribution defined by p in our definition of omniprediction (Definition 2.1).
We show that this not only makes omniprediction possible for constrained problems, but also …

… yield improved runtimes for learning omnipredictors. Our key … This yields improved runtimes
for learning omnipredictors for … We also give efficient omnipredictors when the loss families …

We give the first result for agnostically learning Single-Index Models (SIMs) with arbitrary
monotone and Lipschitz activations. All prior work either held only in the realizable setting or …

… simple construction of omnipredictors for SIMs. We give a learner outputting an omnipredictor
that is … As they are based on Isotron, our omnipredictors are multi-index models with ≈ ε−2 …

We introduce and study the notion of Swap Agnostic Learning. The problem can be phrased
as a game between a* predictor* and an* adversary*: first, the predictor selects a …

… Conceptually, our technical approach for online omniprediction is most closely related to the
… 23] which illustrates a connection between outcome indistinguishability and omniprediction …

… In this paper, we propose the omnipredictor that predicts conditional branches, memory …
conditional predictor along with memory dependency prediction, forming the omnipredictor. …