Slow concept drift is a ubiquitous, yet under-studied problem in practical machine learning
systems. Although recent data is more indicative of future data in these settings, naively …

The most common settings in machine learning to study multitask learning assume either
that a random task is selected on each training trial, or that one task is trained to mastery and …

Training large models with millions or even billions of parameters from scratch incurs
substantial computational costs. Parameter Efficient Fine-Tuning (PEFT) methods …

Many online-learning domains in artificial intelligence involve data with nonstationarities
spanning a wide range of timescales. Heuristic approaches to combat nonstationarity …

We define an evolving in-time Bayesian neural network called a Hidden Markov Neural
Network, which addresses the crucial challenge in time-series forecasting and continual …