Discretization of Series of Communication Signals in Noisy Environment by Reinforcement Learning
Thinking about the “Symbol Grounding Problem” and the brain structure of living things, the author believes that it is the best solution for generating communication in robot-like systems to use a neural network that is trained based on reinforcement learning. As the first step of the research of symbol emergence using neural network, it was examined that parallel analog communication signals are binarized in some degree by noise addition in reinforcement learning-based communication acquisition. In this paper, it is shown that two consecutive analog communication signals are binarized by noise addition using recurrent neural networks. Furthermore, when the noise ratio becomes larger, the degree of the binarization becomes larger.
KeywordsNoise Level Reinforcement Learning Learning Phase Recurrent Neural Network Noise Factor
Unable to display preview. Download preview PDF.
- Sperry, R.W. (1968) Hemisphere Deconnection and Unity in Conscious Awareness, American Psychologist, 23, pp. 723–733Google Scholar
- Shibata, K. and Sugisaka, M. (2004) Discretization of Analog Communication Signals by Noise Addition in Communication Learning, Proc. of AROB 9th, 2, pp. 351–354Google Scholar
- Shibata, K (2004) Discretization of Analog Communication Signals by Noise Addition in Reinforcement Learning of Communication, Technical Report of lEICE, 103(734), NC2003-203, pp. 55–60 (in Japanese)Google Scholar
- Ono, N, Ohira, T. and Rahmani, A.T. (1995) Emergent Organization of Interspecies Communication in Q-Learning Artificial Organs, Advances in Artificial Life, pp.396–405Google Scholar