Towards Making Effective Machine Learning Decisions Against Out-of-Distribution Data

Conventional machine learning systems operate on the assumption of independent and identical distribution (i.i.d), where both the training and test data share a similar sample space, and no distribution shift exists between them. However, this assumption does not hold in practical deployment scenarios, making it crucial to develop methodologies that address the non-trivial task of data distribution shift. In our research, we aim to address this problem by developing ML algorithms that explicitly achieve promising performance when subjected to various types of out-of-distribution (OOD) data. Specifically, we approach the problem by categorizing the data distribution shifts into two types: covariate shifts and semantic shifts, and proposing effective methodologies to tackle each type independently and conjointly while validating them with different types of datasets. We aim to propose ideas that are compatible with existing deep neural networks to perform detection and/or generalization of the test instances that are shifted in semantic and covariate space, respectively.