A novel Continuous Action-set Learning Automaton algorithm

Author(s):  
Ying Guo ◽  
Hao Ge ◽  
Yan Yan ◽  
Yuyang Huang ◽  
Shenghong Li
Author(s):  
Junpei Zhong ◽  
Cornelius Weber ◽  
Stefan Wermter

AbstractRobots and living beings exhibit latencies in their sensorimotor processing due to mechanical and electronic or neural processing delays. A reaction typically occurs to input stimuli of the past. This is critical not only when the environment changes (e.g. moving objects) but also when the agent itself moves. An agent that does not predict while moving may need to remain static between sensory input acquisition and output response to guarantee that the response is appropriate to the percept. We propose a biologically-inspired learning model of predictive sensorimotor integration to compensate for this latency. In this model, an Elman network is developed for sensory prediction and sensory filtering; a Continuous Actor-Critic Learning Automaton (CACLA) is trained for continuous action generation. For a robot docking experiment, this architecture improves the smoothness of the robot’s sensory input and therefore results in a faster and more accurate continuous approach behavior.


2020 ◽  
pp. 1-15
Author(s):  
ALEXANDER S. KECHRIS ◽  
MACIEJ MALICKI ◽  
ARISTOTELIS PANAGIOTOPOULOS ◽  
JOSEPH ZIELINSKI

Abstract It is a long-standing open question whether every Polish group that is not locally compact admits a Borel action on a standard Borel space whose associated orbit equivalence relation is not essentially countable. We answer this question positively for the class of all Polish groups that embed in the isometry group of a locally compact metric space. This class contains all non-archimedean Polish groups, for which we provide an alternative proof based on a new criterion for non-essential countability. Finally, we provide the following variant of a theorem of Solecki: every infinite-dimensional Banach space has a continuous action whose orbit equivalence relation is Borel but not essentially countable.


Author(s):  
Yuntao Han ◽  
Qibin Zhou ◽  
Fuqing Duan

AbstractThe digital curling game is a two-player zero-sum extensive game in a continuous action space. There are some challenging problems that are still not solved well, such as the uncertainty of strategy, the large game tree searching, and the use of large amounts of supervised data, etc. In this work, we combine NFSP and KR-UCT for digital curling games, where NFSP uses two adversary learning networks and can automatically produce supervised data, and KR-UCT can be used for large game tree searching in continuous action space. We propose two reward mechanisms to make reinforcement learning converge quickly. Experimental results validate the proposed method, and show the strategy model can reach the Nash equilibrium.


2018 ◽  
Vol 89 ◽  
pp. 68-81 ◽  
Author(s):  
Túlio A.M. Toffolo ◽  
Jan Christiaens ◽  
Sam Van Malderen ◽  
Tony Wauters ◽  
Greet Vanden Berghe

Author(s):  
A. V. Fominykh ◽  
I. N. Mikolaychik ◽  
Yu. I. Ovchinnikova ◽  
D. N. Ovchinnikov

Russian market of complete feed recent years shows a steady development. Currently the Kurgan region the state of base of feed and a shortage of specialized feed additives leads to a violation of maintenance conditions for livestock, reduced productivity. The increase in production of concentrates and complete feed currently is an actual scientific and technical problem whose solution is of great scientific and practical importance. To solve this problem, the technology of production of animal feed and PMVA on the basis of full-fat extruded soybeans produced using a shaft cooler and sapropel dehydrated with the use of filtering centrifuges continuous action has been proposed. The proposed technology of production of animal feed and PMVA completely solves the problem of removing excess moisture from the sapropel immediately after the production and delivery of product to the shore. The process of extraction and preparation for use of sapropel reduced from several days or months (winter freezing) up to several hours.


Sign in / Sign up

Export Citation Format

Share Document