Binary Space Partitioning as Intrinsic Reward (Skaba)

Added by Deon Garrett about 7 years ago

An autonomous agent embodied in a humanoid robot, in order to learn from the overwhelming flow of raw and noisy sensory, has to effectively reduce the high spatial-temporal data dimensionality. In this paper we propose a novel method of unsupervised feature extraction and selection with binary space partitioning, followed by a computation of information gain that is interpreted as intrinsic reward, then applied as immediate-reward signal for the reinforcement-learning. The space partitioning is executed by tiny codelets running on a simulated Turing Machine. The features are represented by concept nodes arranged in a hierarchy, in which those of a lower level become the input vectors of a higher level.

paper_43.pdf (195.8 kB)

Replies (1)

RE: Binary Space Partitioning as Intrinsic Reward (Skaba) - Added by Drew Britt over 1 year ago

All the chapters have been assigned for the help of the people. The right use of the [[]] is improved for the good and all refined norms or the humans.