Meta-Learning surrogate models for sequential decision making
This provides a general black-box learning approach for sequential decision making problems, though it appears incremental as it builds on existing meta-learning and probabilistic modeling concepts.
The authors tackled sequential decision making problems by introducing a unified probabilistic framework using meta-learning surrogate models, which demonstrated efficient adaptation across diverse domains including control problems, recommender systems, and adversarial attacks on RL agents.
We introduce a unified probabilistic framework for solving sequential decision making problems ranging from Bayesian optimisation to contextual bandits and reinforcement learning. This is accomplished by a probabilistic model-based approach that explains observed data while capturing predictive uncertainty during the decision making process. Crucially, this probabilistic model is chosen to be a Meta-Learning system that allows learning from a distribution of related problems, allowing data efficient adaptation to a target task. As a suitable instantiation of this framework, we explore the use of Neural processes due to statistical and computational desiderata. We apply our framework to a broad range of problem domains, such as control problems, recommender systems and adversarial attacks on RL agents, demonstrating an efficient and general black-box learning approach.