Fairness Hub Technical Briefs: Definition and Detection of Distribution Shift
This work addresses distribution shift issues for practitioners in educational machine learning applications, but it appears incremental as it focuses on definition and detection without introducing new methods.
The paper tackles the problem of distribution shift in machine learning, where training and real-world data differ, leading to performance reductions, and focuses on defining and detecting these shifts in educational settings for standard prediction tasks.
Distribution shift is a common situation in machine learning tasks, where the data used for training a model is different from the data the model is applied to in the real world. This issue arises across multiple technical settings: from standard prediction tasks, to time-series forecasting, and to more recent applications of large language models (LLMs). This mismatch can lead to performance reductions, and can be related to a multiplicity of factors: sampling issues and non-representative data, changes in the environment or policies, or the emergence of previously unseen scenarios. This brief focuses on the definition and detection of distribution shifts in educational settings. We focus on standard prediction problems, where the task is to learn a model that takes in a series of input (predictors) $X=(x_1,x_2,...,x_m)$ and produces an output $Y=f(X)$.