A Roadmap towards Machine Intelligence
This work addresses the foundational problem of machine intelligence for AI researchers, but it is incremental as it builds on existing ideas without presenting new experimental results.
The paper tackles the challenge of developing intelligent machines by proposing fundamental properties they should have, focusing on communication and learning, and discusses a simple environment for teaching natural-language-based communication as a prerequisite for complex human interaction.
The development of intelligent machines is one of the biggest unsolved challenges in computer science. In this paper, we propose some fundamental properties these machines should have, focusing in particular on communication and learning. We discuss a simple environment that could be used to incrementally teach a machine the basics of natural-language-based communication, as a prerequisite to more complex interaction with human users. We also present some conjectures on the sort of algorithms the machine should support in order to profitably learn from the environment.