Multi-layer in-place learning for autonomous mental development
Cortical self-organization during open-ended development is a core issue for perceptual development. Traditionally, unsupervised learning and supervised learning are two different types of learning conducted by different networks. However, there is no evidence that the biological nervous system treats them in a disintegrated way. The computational model presented here integrates both types of learning using a new biologically inspired network whose learning is in place. By in-place learning, we mean that each neuron in the network learns on its own while interacting with other neurons. There is no need for a separate learning network. Presented in this thesis is the Multi-layer In-place Learning Network (MILN) for regression and classification. This work concentrates on its two-layer version (without incorporating an attention selection mechanism). The network enables both unsupervised and supervised learning to occur concurrently. Within each layer, the adaptation of each neuron is nearly-optimal in the sense of the least possible estimation error given the observations. MILN is intended to provide memory organization capability and to be used as a general-purpose regressor in Autonomous Mental Development. The theory behind the network is discussed in detail. Experimental results are presented to show the effects of the properties investigated.