This paper describes how agents can learn an internal model of the world structurally by focusing on the problem of behavior-based articulation. We develop an on-line learning scheme -- the so-called mixture of recurrent neural net (RNN) experts -- in which a set of RNN modules becomes self-organized as experts on multiple levels in order to account for the different categories of sensory-motor flow which the robot experiences. Autonomous switching of activated modules in the lower level actually represents the articulation of the sensory-motor flow. In the meanwhile, a set of RNNs in the higher level competes to learn the sequences of module switching in the lower level, by which articulation at a further more abstract level can be achieved. The proposed scheme was examined through simulation experiments involving the navigation learning problem. Our dynamical systems analysis clarified the mechanism of the articulation; the possible correspondence between the articulation...

The paper examines characteristics of interactive learning between human tutors and a robot having a dynamic neural network model which is inspired by human parietal cortex functions. A humanoid robot, with a recurrent neural network that has a hierarchical structure, learns to manipulate objects. Robots learn tasks in repeated self-trials with the assistance of human interaction which provides physical guidance until tasks are mastered and learning is consolidated within neural networks. Experimental results and the analyses showed that 1) codevelopmental shaping of task behaviors stems from interactions between the robot and tutor, 2) dynamic structures for articulating and sequencing of behavior primitives are selforganized in the hierarchically organized network, and 3) such structures can afford both generalization and context-dependency in generating skilled behaviors.

Behavior may be controlled by reactive systems. In a reactive system the motor output is exclusively driven by actual sensory input. An alternative solution to control behavior is given by “cognitive ” systems capable of planning ahead. To this end the system has to be equipped with some kind of internal world model. A sensible basis of an internal world model might be a model of the system’s own body. I show that a reactive system with the ability to control a body of complex geometry requires only a slight reorganization to form a cognitive system. This implies that the assumption that the evolution of cognitive properties requires the introduction of new, additional modules, namely internal world models, is not justified. Rather, these modules may already have existed before the system obtained cognitive properties. Furthermore, I discuss whether the occurrence of such world models may lead to systems having internal perspective.

Abstract not found

In this p#isW , we investigated interactive learning between human subjects and robotexp#16779W-508 , and its essential characteristics are examined using the dynamical systems ap#stemsW Our research concentrated on the navigation system of a sp#1140W- develop#- humanoid robot called Robovie and seven human subjects whose eyes were covered, making themdep#W1397 on the robot for directions. Wecomp#2 ed the usual feed-forward neural network (FFNN) without recursive connections and the recurrent neural network (RNN). Although thep#W20724W-41 obtained with both the RNN and the FFNNimp#W206 in the early stages of learning, as the subject changed theop#W24760 by learning on its own, allp#W16238W-37 gradually became unstable and failed. Results of a questionnaire given to the subjects confirmed that the FFNN gives better mental imp ressions,esp#s,W103 from theasp#18 ofop#2557W-34 . When the robot used a consolidation-learning algorithm using the rehearsaloutp#sa of the RNN, thep#W20908W-3 imp##090 even when interactive learning continued for a long time. The questionnaire results then also confirmed that the subject's mentalimp# essions of the RNN imp# oved significantly. The dynamical systems analysis of RNNssup#W2 t these differences.

I am most grateful to all those who commented on ‘“The Self”’. The result was a festival of misunderstanding, but misunderstanding is one of the great engines of progress. Few of the contributors to the symposium on ‘Models of the Self ’ were interested in my project: some (like Olson and Wilkes) were already highly sceptical

Dynamical systems theory provides effective formal models of structure in natural languages. We describe a recurrent neural network called the Bramble Network (BRN) and a related analytical tool, the Dynamical Automaton (DA), which process words in sequence. The BRN makes one discrete jump across its state space each time a word is processed, and then settles continuously to a stable state. Processing time is modeled as convergence time. Two well-known phenomena in natural language processing are modeled: (i) the inverse correlation between word frequency and response time and (ii) the correlation between parsing difficulty and level of center embedding. The model shows how constructs of dynamical systems theory provide a link between neural network models which are good at learning and show human-like flexibility and abstract linguistic representations which are the current best model of natural language syntactic structure and interpretation. What role does dyna

In this paper, we will explore possibilities of dynamic interactions between human and neurocognitive robots especially focusing on the psychological problems of joint attentions and turn-taking. Firstly, we will show that movement patterns of a joystick-type haptic device which are driven by a simple attractor-based memory dynamics of recurrent neural network (RNN) can introduce novel interactive experiences to human subjects based on their force and proprioceptional sensations. Secondly, we will show experiments of joint attention game between human and a humanoid robot based on imitation learning. In the experiments, an extended scheme of RNNs is utilized for constructing a mirror system by which recognition of other’s movements and generation of owns can be naturally synchronized in the realtime imitation. These experiments suggest that spontaneous shifts in joint attentions as well as turn taking have resulted from so-called the open-dynamic structures where stable and unstable manifold coexist in the coupling between the robots and human cognitive processes.

We study the diversity of complex spatio-temporal patterns in the behavior of random synchronous asymmetric neural networks (RSANNs). Special attention is given to the impact of disordered threshold values on limit-cycle diversity and limit-cycle complexity in RSANNs which have ‘normal ’ thresholds by default. Surprisingly, RSANNs exhibit only a small repertoire of rather complex limit-cycle patterns when all parameters are fixed. This repertoire of complex patterns is also rather stable with respect to small parameter changes. These two unexpected results may generalize to the study of other complex systems. In order to reach beyond this seemingly-disabling ‘stable and small ’ aspect of the limit-cycle repertoire of RSANNs, we have found that if an RSANN has threshold disorder above a critical level, then there is a rapid increase of the size of the repertoire of patterns. The repertoire size initially follows a power-law function of the magnitude of the threshold disorder. As the disorder increases further, the limit-cycle patterns themselves become simpler until at a second critical level most of the limit cycles become simple fixed points. Nonetheless, for moderate changes in the threshold parameters, RSANNs are found to display specific features of behavior desired for rapidly-responding processing

The current article suggests that deterministic chaos self-organized in cortical dynamics could be responsible for the generation of spontaneous action sequences. Recently, various psychological observations have suggested that humans and primates can learn to extract statistical structures hidden in perceptual sequences experienced during active environmental interactions. Although it has been suggested that such statistical structures involve chunking or compositional primitives, their neuronal implementations in brains have not yet been clarified. Therefore, to reconstruct the phenomena, synthetic neuro-robotics experiments were conducted by using a neural network model, which is characterized by a generative model with intentional states and its multiple timescales dynamics. The experimental results showed that the robot successfully learned to imitate tutored behavioral sequence patterns by extracting the underlying transition probability among primitive actions. An analysis revealed that a set of primitive action patterns was embedded in the fast dynamics part, and the chaotic dynamics of spontaneously sequencing these action primitive patterns was structured in the slow dynamics part, provided that the timescale was adequately set for each part. It was also shown that self-organization of this type of functional hierarchy ensured robust action generation by the robot in its interactions with a noisy environment. This article discusses the correspondence of the synthetic experiments with the known hierarchy of the prefrontal cortex, the supplementary motor area, and the primary motor cortex for action generation. We speculate