A no-free-lunch theorem for multitask learning

DOI10.1214/22-AOS2189MaRDI QIDQ2112800

Samory Kpotufe, Steve Hanneke

Publication date: 12 January 2023

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2006.15785

zbMATH Keywords

classificationstatistical learning theorytransfer learningmultitask learning

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Computational learning theory (68Q32) Learning and adaptive systems in artificial intelligence (68T05) Pattern recognition, speech recognition (68T10)

Related Items (max. 100)

Transfer Learning in Large-Scale Gaussian Graphical Models with False Discovery Rate Control ⋮ Adaptive and robust multi-task learning ⋮ Transfer Learning under High-dimensional Generalized Linear Models ⋮ Statistical learning from biased training samples

Cites Work

• Oracle inequalities and optimal inference under group sparsity
• Risk bounds for statistical learning
• A notion of task relatedness yielding provable multiple-task learning guarantees
• Distribution inequalities for the binomial law
• A Bayesian/information theoretic model of learning to learn via multiple task sampling
• On the complexity of learning from drifting distributions
• Concentration around the mean for maxima of empirical processes
• Optimal aggregation of classifiers in statistical learning.
• A theory of transfer learning with applications to active learning
• A theory of learning from different domains
• Marginal singularity and the benefits of labels in covariate-shift
• Adaptive transfer learning
• Local Rademacher complexities and oracle inequalities in risk minimization. (2004 IMS Medallion Lecture). (With discussions and rejoinder)
• Transfer learning for nonparametric classification: minimax rate and adaptive classifier
• New Analysis and Algorithm for Learning with Drifting Distributions
• Sample Selection Bias Correction Theory
• The information complexity of learning tasks, their structure and their distance
• Simultaneous Support Recovery in High Dimensions: Benefits and Perils of Block $\ell _{1}/\ell _{\infty} $-Regularization
• Convexity, Classification, and Risk Bounds
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item

This page was built for publication: A no-free-lunch theorem for multitask learning