Learn to Expect the Unexpected: Probably Approximately Correct Domain Generalization
AISTATS, pp. 3574-3582, 2021.
Domain generalization is the problem of machine learning when the training data and the test data come from different data domains. We present a simple theoretical model of learning to generalize across domains in which there is a meta-distribution over data distributions, and those data distributions may even have different supports. I...More
PPT (Upload PPT)