Robust Probabilistic Modeling with Bayesian Data Reweighting
This addresses robustness issues in probabilistic modeling for applications like recommendation systems, though it appears incremental as an extension of existing reweighting techniques.
The paper tackles the problem of data that deviates from probabilistic model assumptions by proposing a method to detect and mitigate mismatch, which improves predictive accuracy as demonstrated on the Movielens 1M dataset.
Probabilistic models analyze data by relying on a set of assumptions. Data that exhibit deviations from these assumptions can undermine inference and prediction quality. Robust models offer protection against mismatch between a model's assumptions and reality. We propose a way to systematically detect and mitigate mismatch of a large class of probabilistic models. The idea is to raise the likelihood of each observation to a weight and then to infer both the latent variables and the weights from data. Inferring the weights allows a model to identify observations that match its assumptions and down-weight others. This enables robust inference and improves predictive accuracy. We study four different forms of mismatch with reality, ranging from missing latent groups to structure misspecification. A Poisson factorization analysis of the Movielens 1M dataset shows the benefits of this approach in a practical scenario.