"4981","Covert Perceptual Capability Development","In this paper, we propose a model to develop robots’ covert perceptual capability using reinforcement learning. Covert perceptual behavior is treated as action selected by a motivational system. We apply this model to vision-based navigation. The goal is to enable a robot to learn road boundary type. Instead of dealing with problems in controlled environments with a low-dimensional state space, we test the model on images captured in non-stationary environments. Incremental Hierarchical Discriminant Regression is used to generate states on the fly. Its coarse-to-fine tree structure guarantees real-time retrieval in high-dimensional state space. K Nearest-Neighbor strategy is adopted to further reduce training time complexity.","http://cogprints.org/4981/","Huang, Xiao and Weng, Juyang","Berthouze, Luc and Kaplan, Frédéric and Kozima, Hideki and Yano, Hiroyuki and Konczak, Jürgen and Metta, Giorgio and Nadel, Jacqueline and Sandini, Giulio and Stojanov, Georgi and Balkenius, Christian","       <span class=\"person_name\">Huang, Xiao</span> and <span class=\"person_name\">Weng, Juyang</span>     (2005)  <a href=\"http://cogprints.org/4981/\"><em>Covert Perceptual Capability Development.</em></a>       [Conference Paper]           ","","2005"