Brief

Researchers have introduced a new combinatorial measure called the domain shattering dimension to address a core question in domain generalization. This measure quantifies how many randomly sampled domains are needed to train a model that performs well across all domains within a given family. The study establishes a tight relationship between this new dimension and the classic VC dimension, proving that any hypothesis class learnable in the standard PAC framework is also learnable in this domain generalization context. AI

Researchers have developed a new framework for incentivized collaboration in active learning, where multiple agents work together to label data while minimizing costs. The proposed protocols ensure that individual agents cannot improve their outcomes by acting alone. While computing the optimal algorithm is computationally difficult, the new protocols offer a practical approach that is comparable to existing approximation algorithms in terms of label complexity. AI