20 results for βClass probabilitiesβ
CS papers onlyHybrid search: Keyword + semantic, ranked by combined score.β
Want pure semantic search? Try claim verification β
This paper establishes a large deviation principle for the generalization error of interpolating classifiers in the overparametrized regime.
Computes class probabilities in block-independent disjoint probabilistic databases.
Li Zhang, Yuyuan Li, XiaoHua Feng, Jiaming Zhang +2 more
This paper addresses the challenge of achieving optimal fairness and accuracy simultaneously in multi-class classification by proposing novel in-processing and post-processing algorithms that convergeβ¦
The paper proves that the majority vote of three independent consistent classifiers is an optimal learner in the realizable PAC setting.
This paper characterizes proper binary classification from positive-only samples, revealing a rich landscape that differs from standard PAC learning.
This paper provides non-asymptotic and explicit estimates for the exponential deviation inequalities of the HyperLogLog estimator.
Zakk Heile, Hayden McTavish, Varun Babbar, Margo Seltzer +1 more
The paper introduces PRAXIS, a novel algorithm that efficiently approximates the computation of 'Rashomon sets' for decision trees, significantly reducing memory and runtime complexity.
This paper introduces classes of functions that compute the k-th solution for certain types of problems and proves their inter-reducibility with corresponding classes for different order functions. Itβ¦
This paper formalizes test-time training (TTT) as a problem of sampling from a given probability measure and shows a quadratic lower bound on the query complexity of sampling.
Zexin Li, Weibo Zhang, Linyi Li, Arvind Easwaran +1 more
A query-efficient framework for black-box probabilistic verification.
This paper provides an information-theoretic analysis of Yager's negation of probability distributions and its generalizations.
The paper extends the minimum-distance estimator approach for density estimation from total variation distance to Hellinger distance.
This paper extends gradient boosting to functions of vector inputs using a simple algorithm with histogram-based decision trees.
The authors prove a new collapse in the polynomial hierarchy and introduce a new class of languages.
A new boosting algorithm that strong learns concept classes closed under O(log 1/Ξ³)-XOR using O(log 1/Ξ΅) calls to a Ξ³-advantage weak learner and additional samples, by connecting boosting with list-deβ¦