Detecting and Adapting to Irregular Distribution Shifts in Bayesian Online Learning
This work is significant for practitioners in online learning who need to maintain model performance in environments with unpredictable concept or covariate drifts.
This paper addresses online learning under irregular distribution shifts by developing a Bayesian online inference approach. It simultaneously infers shifts and adapts the model by partially erasing past information upon change detection, outperforming state-of-the-art Bayesian online learning methods.
We consider the problem of online learning in the presence of distribution shifts that occur at an unknown rate and of unknown intensity. We derive a new Bayesian online inference approach to simultaneously infer these distribution shifts and adapt the model to the detected changes by integrating ideas from change point detection, switching dynamical systems, and Bayesian online learning. Using a binary 'change variable,' we construct an informative prior such that--if a change is detected--the model partially erases the information of past model updates by tempering to facilitate adaptation to the new data distribution. Furthermore, the approach uses beam search to track multiple change-point hypotheses and selects the most probable one in hindsight. Our proposed method is model-agnostic, applicable in both supervised and unsupervised learning settings, suitable for an environment of concept drifts or covariate drifts, and yields improvements over state-of-the-art Bayesian online learning approaches.