Long-term learning of concept and word by robots: Interactive learning framework and preliminary results

Abstract

One of the biggest challenges in intelligent robotics is to build robots that can understand and use language. Such robots will be a part of our everyday life; at the same time, they can be of great help to investigate the complex mechanism of language acquisition by infants in constructive approach. To this end, we think that the practical long-term on-line concept/word learning algorithm for robots and the interactive learning framework are the key issues to be addressed. In this paper we develop a practical on-line learning algorithm that solves three remaining problems in our previous study. We also propose an interactive learning framework, in which the proposed on-line learning algorithm is embedded. The main contribution of this paper is to develop such a practical learning framework, and we test it on a real robot platform to show its potential toward the ultimate goal.

Online Algorithm for Robots to Learn Object Concepts and Language Model

“…To accomplish this, robots would need a model to infer unobserved information using the generated concepts, i.e., Please note that the "Feedback Loop" represented by the thick allow in the figure is the main difference between our proposal and [6], [7]. This feedback loop significantly improves learning results, as shown in later experiments.…”

“…First, in a rigorous manner, we formulate the online joint learning of concepts and a language model, based on a generative model. In contrast to this paper, [6] and [7] provide no theoretical formulation of the joint learning problem, and the online algorithm is not involved in [9]. Hence, this is the first attempt to propose an online joint learning algorithm that achieves the aforementioned learning objectives, as shown in Fig.…”

“…At the recognition stage, the trained model can be used to infer unobserved information from observed information. Since the original MLDA encounters problems with batch learning, it was extended to an online version in [6], [7]. A batch-type algorithm normally assumes that the system can immediately use all available training data.…”

“…Therefore, an algorithm that can learn more quickly is desired. To tackle this problem, we have in past work proposed a "Particle Filter-based online MLDA" (PFoMLDA), an online extension of MLDA [6], [7]. PFoMLDA allows robots to incrementally form object concepts in real environments, and is the baseline algorithm used in this paper.…”

“…This causes phoneme recognition errors that significantly affect the learning of concepts. Reference [7] reported a degradation in learning performance due to phoneme recognition errors. As is well known, the performance of recent speech recognition technology heavily depends on the language model.…”