Predictive representations: building blocks of intelligence
This work addresses the problem of understanding and modeling intelligence for researchers in AI, cognition, and neuroscience, but it is incremental as it builds on existing theoretical ideas.
The paper integrates reinforcement learning theory with cognitive and neuroscience research, focusing on the successor representation and its generalizations as predictive representations that may serve as versatile building blocks for intelligence.
Adaptive behavior often requires predicting future events. The theory of reinforcement learning prescribes what kinds of predictive representations are useful and how to compute them. This paper integrates these theoretical ideas with work on cognition and neuroscience. We pay special attention to the successor representation (SR) and its generalizations, which have been widely applied both as engineering tools and models of brain function. This convergence suggests that particular kinds of predictive representations may function as versatile building blocks of intelligence.