Thisallows agents to directly learn in an end-to-end manner to utilise relevant information to inform their outputs. In addition, new information can be attended to by the agent, without retraining, by simply augmenting the retrieval dataset.

Thisallows agents to directly learn in an end-to-end manner to utilise relevant information to inform their outputs. In addition, new information can be attended to by the agent, without retraining, by simply augmenting the retrieval dataset.

Work done as a visiting student at MIT. 38th Conference on Neural Information Processing Systems (NeurIPS 2024).

Nevertheless, theperformance guarantee of adaptive TD with neural network approximation remains widely unknown.

Our approach relies on the principle ofoptimism under uncertaintyto efficiently explore. Our learning algorithms optimizetheiractions withrespect toamodel based ontheempirical statistics, while optimistically overestimating rewards and underestimating the resource consumption (i.e., overestimating the distance from the constraint).