Experimental and causal view on information integration in autonomous agents
This work addresses the challenge of information integration for autonomous agents like self-driving cars, but it is incremental as it presents preliminary ideas and simple implementations.
The paper tackles the problem of integrating heterogeneous information for autonomous agents by exploring how experiments and causal information can improve understanding and data-efficient decision-making, presenting a simple agent that integrates video data and evaluates it on the Malmo platform.
The amount of digitally available but heterogeneous information about the world is remarkable, and new technologies such as self-driving cars, smart homes, or the internet of things may further increase it. In this paper we present preliminary ideas about certain aspects of the problem of how such heterogeneous information can be harnessed by autonomous agents. After discussing potentials and limitations of some existing approaches, we investigate how \emph{experiments} can help to obtain a better understanding of the problem. Specifically, we present a simple agent that integrates video data from a different agent, and implement and evaluate a version of it on the novel experimentation platform \emph{Malmo}. The focus of a second investigation is on how information about the hardware of different agents, the agents' sensory data, and \emph{causal} information can be utilized for knowledge transfer between agents and subsequently more data-efficient decision making. Finally, we discuss potential future steps w.r.t.\ theory and experimentation, and formulate open questions.