Pretraining a language model (LM) on text has been shown to help various downstream NLP tasks.

Consider a bandit algorithm that recommends actions to self-interested users in a recommendation system. The users are free to choose other actions and need to be incentivized to follow the algorithm's recommendations.

Sequential data in our everyday life is often hierarchically structured.

Imitation learning aims to extract high-performance policies from logged demonstrations of expert behavior.

One approach toward automatic grading is to learn an agent that interacts with a student's program and explores states indicative of errors via reinforcement learning. However, existing work on this approach only provides binary feedback of whether a program is correct or not, while students require finer-grained feedback on the specific errors in their programs to understand their mistakes. In this work, we show that exploring to discover errors can be cast as a meta-exploration problem.

While deep learning's performance has led it to become the dominant paradigm in machine learning,

We extensively validate our method on both label-and domain-heterogeneous settings, on which it outperforms the state-of-the-art personalized federated learning methods.

We prove that a variant of EW A either achieves a negative regret (i.e., the algorithm