Towards Information-Seeking AgentsDownload PDF

28 Mar 2024 (modified: 21 Jul 2022)Submitted to ICLR 2017Readers: Everyone
Abstract: We develop a general problem setting for training and testing the ability of agents to gather information efficiently. Specifically, we present a collection of tasks in which success requires searching through a partially-observed environment, for fragments of information which can be pieced together to accomplish various goals. We combine deep architectures with techniques from reinforcement learning to develop agents that solve our tasks. We shape the behavior of these agents by combining extrinsic and intrinsic rewards. We empirically demonstrate that these agents learn to search actively and intelligently for new information to reduce their uncertainty, and to exploit information they have already acquired.
TL;DR: We investigate the behavior of models trained to answer questions by asking sequences of simple questions.
Conflicts: maluuba.com, mcgill.ca, umontreal.ca
7 Replies

Loading