A hand-designed internal representation of the world cannot deal with unknown or uncontrolled environments. Motivated by human cognitive and behavioral development, this paper presents a theory, an architecture, and some experimental results for developmental robotics. By a developmental robot, we mean that the robot generates its “brain ” (or “central nervous system, ” including the information processor and controller) through online, real-time interactions with its environment (including humans). A new Self-Aware Self-Effecting (SASE) agent concept is proposed, based on our SAIL and Dav developmental robots. The manual and autonomous development paradigms are formulated along with a theory of representation suited for autonomous development. Unlike traditional robot learning, the tasks that a developmental robot ends up learning are unknown during the programming time so that the task-specific representation must be generated and updated through real-time “living ” experiences. Experimental results with SAIL and Dav developmental robots are presented, including visual attention selection, autonomous navigation, developmental speech learning, range-based obstacle avoidance, and scaffolding through transfer and chaining.

Abstract—Recently, it has been shown that top-down connections improve recognition in supervised learning. In the work presented here, we show how top-down connections represent temporal context as expectation and how such expectation assists perception in a continuously changing physical world, with which an agent interacts during its developmental learning. In experiments in object recognition and vehicle recognition using two types of networks (which derive either global or local features), it is shown how expectation greatly improves performance, to nearly 100 % after the transition periods. We also analyze why expectation will improve performance in such real world contexts. I.

This paper presents a computational theory of developmental mental architectures for artificial and natural systems, motivated by neuroscience. The work is an attempt to approximately model biological mental architectures using mathematical tools. Six types of architecture are presented, beginning with the observation-driven Markov decision process as Type-1. From Type-1 to Type-6, the architecture progressively becomes more complete toward the necessary functions of autonomous mental development. Properties of each type are presented. Experiments are discussed with emphasis on their architectures. r 2007 Published by Elsevier B.V.

This paper proposes and implements a new experimental paradigm to study the role of multimodal interaction in automatic language learning. We observe that child language acquisition relies significantly on everyday social interactions with adult partners. In light of this, we argue that an important step to build machines that can also learn from social interactions with human users is to understand the nature of learning-oriented interactions. To do so, a central problem is to find a way to decouple the social interaction between two agents (e.g. a human supervisor and a machine learner), so that we can systematically manipulate and control the dynamic flow of the interaction to create and examine various interactive learning conditions. In this paper, we build a set of virtual humans as language learners possessing different social-cognitive skills, and ask real people to teach them object names. Using multisensory recording devices, we measured how well real people interact with virtual humans and how they shape their behaviors to adapt to different social-cognitive skills that virtual humans possess. Multimodal data were analyzed to shed light on both perceptual and behavioral aspects of human users in interaction. These results can be used both to guide building artificially intelligent system and to provide useful insights on humanhuman communication and child language learning.