2011/12/06 by Joseph Modayil, Adam White, Modayil, Joseph +3
Computer Science · Neuroscience · #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Neural dynamics and brain function #Reinforcement Learning in Robotics #Robotics (cs.RO) #cs.LG #cs.RO
paper · pdf · doi:10.48550/arxiv.1112.1133
(11 pages, 5 figures, This version to appear in the Proceedings of the Conference on the Simulation of Adaptive Behavior, 2012)
openalex publication_date 2011/12/06 · arxiv created 2012/06/08 · arxiv updated 2015/03/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The term "nexting" has been used by psychologists to refer to the propensity of people and many other animals to continually predict what will happen next in an immediate, local, and personal sense. The ability to "next" constitutes a basic kind of awareness and knowledge of one's environment. In this paper we present results with a robot that learns to next in real time, predicting thousands of features of the world's state, including all sensory inputs, at timescales from 0.1 to 8 seconds. This was achieved by treating each state feature as a reward-like target and applying temporal-difference methods to learn a corresponding value function with a discount rate corresponding to the timescale. We show that two thousand predictions, each dependent on six thousand state features, can be learned and updated online at better than 10Hz on a laptop computer, using the standard TD(lambda) algorithm with linear function approximation. We show that this approach is efficient enough to be practical, with most of the learning complete within 30 minutes. We also show that a single tile-coded feature representation suffices to accurately predict many different signals at a significant range of timescales. Finally, we show that the accuracy of our learned predictions compares favorably with the optimal off-line solution.