Towards Information-Seeking Agents
This work addresses the challenge of information-seeking for AI agents, though it appears incremental as it builds on existing deep learning and reinforcement learning techniques.
The paper tackles the problem of training agents to efficiently gather and piece together information in partially-observed environments, resulting in agents that learn to actively search for new information and exploit acquired knowledge.
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.