The Role of Sensorimotor Function, Associative Memory and Reinforcement Learning in Automatic Acquisition of Spoken Language by an Autonomous Robot

摘要

We propose to test experimentally a novel theory of cognition by building an autonomous intelligent robot. Our theory comprises three equally important principles. First, we submit that cognition, as manifest in humans, requires a well-integrated sensorimotor periphery. For the purposes of this experiment sensorimotor function includes binaural audio, stereo video, tactile sense, and proprioceptive control of motion and manipulation of objects. Our methods are designed to exploit the synergy intrinsic in the combined sensorimotor signals. This sensory fusion is essential for the development of a semantic representation of reality from which all other levels of linguistic structure derive. Second, for intelligent behavior, the widely accepted role of computation in the sense of Turing is of secondary significance to the primary mechanism of associative memory. We plan to experiment with several different architectures of such a memory. And, finally, the contents of the associative memory must be acquired by the interaction of the machine with the physical world in a reinforcement-training regime. The reinforcement signal is a direct, real-time, on-line evaluation of only the success or failure of the robot's behavior in response to some stimulus. This signal comes from three sources: autonomous experimentation by the robot, instruction of the robot by a benevolent teacher as to the success or failure of its behavior, and instruction of the robot by the teacher in the form of direct physical demonstration of the desired behavior (e. g. overhauling the robot's actuators). Such instruction must make no use of any supervised training based on pre-classified data. Nor may the robot use any predetermined representation of concepts or algorithms. We intend to produce a robot, trained as described above, of sufficient complexity to be able to carry out simple navigation and object manipulation tasks in response to naturally spoken commands. The linguistic competence of the robot is to be acquired along with its other cognitive abilities in the course of its training. This will result from the synergistic effect that the behavior of a complex combination of simple parts can be much richer than would be predicted by analyzing the components in isolation.