Research

Publications

Ioana Goga's PhD research

Coneural > Ioana Goga > PhD research

Educating attention in cognitive robots

One obvious source of inspiration for building integrated, artificial systems that can scale both the size and complexity of their behavioral repertoire is the process of human cognitive development. The rapprochement of developmental and social psychology, robotics, and neuroscience has given birth to several research fields: epigenetic robotics (Zlatev and Balkenius, 2001), cognitive robotics (Weng et al., 2000), social robotics (Billard and Dautenhahn, 2000). In this thesis, we will follow a developmental approach to study the mechanisms by which human infants acquire goal-directed behavior and language and to investigate how to develop similar competences in robots. The focus will be on investigating how interactive teaching, based on tutoring and demonstration can be implemented in robots.

Learning by imitation has many of the desirable characteristics of animal-like, online learning, and endows the robot with the capacity to continually learn new behaviors while observing and interacting with other agents (Demiris and Hayes, 2001). In order to assist the imitator in recognizing the goal of the demonstration and in structuring the imitation behavior, the teacher can use instruction. In learning by instruction the agent is given information about the environment, domain knowledge, or about how to accomplish a particular task on-line. There is recent interest in the field, towards collaborative learning based on joint intention and progressive tutoring of the learner (Breazeal et al., 2004).

The novelty of our approach results from the investigation of the imitation and tutoring mechanisms from an integrative perspective. Both can be seen as building blocks of a general bootstrapping mechanism, which uses all available means to focus attention, extract a bit of knowledge, and use this knowledge to perform little more analysis on future inputs, and thereby, reduce the uncertainty. The model starts with a set of pre-programmed behaviors (i.e., gaze following, skin color preference, visuomotor coordination, grasping abilities) and develops in an incremental manner goal-directed behavior, intentionality and language. We refer to this bootstrapping process based on demonstration and tutoring as a process of educating attention in cognitive robots.

	The seriate nesting cups experiment	Psycholinguistic studies

Testing psychological assumptions	Computational model of the seriate cups task	Computational model of manipulation and linguistic skills	Validation through psychological experiments

	Implementation on a neural network architecture	Implementation and simulation

	Dynamic simulation with robotic agents

Figure 1. My PhD research plan. The yellow activity flow has been accomplished, while the white boxes are under development. Click on any box on the graphic, to see details on the related activities.

Preliminary work has been carried out towards the design of a cognitive architecture, which can support learning by imitation and instruction (Goga and Billard, 2004). An essential component of the cognitive architecture is represented by the attention module, whose function is to direct gaze towards objects of interest in the environment. A two-component framework for attention deployment has been implemented, inspired by recent research in modeling of visual attention (Goga and Billard, 2006).

An important tenet of this thesis research is that it builds on real data from psychological, developmental and neurobiological studies. The starting point is represented by the definition of a developmental benchmark, against which modeling can be compared. A good candidate is the seriated nesting cups task (Greenfield et al., 1972). Preliminary work focused on investigating with modeling and experimental methods the various types of constraints (i.e., social, computational and maturational) involved in the performance of the seriated nesting cups task (Goga and Billard, in press). The outcome of these computer simulation investigations is represented by a developmentally constrained model able to reproduce the limitations observed in the human infants' strategies (Goga and Billard, in press). We intend to refine and extend the action system currently developed for the imitation of the seriated nesting cups task towards the integration with a language system capable to learn from instruction.

Accomplished and expected results:

an original computational model of the developmental stages of seriate cups behavior
video, audio and chat transcriptions of the child-caretaker interaction taking place during the seriate nesting cups task. This data collection can be valuable for any sort of modeling and simulation work which investigates the interplay of manipulatory, linguistic and imitative skills in humanoids
a biologically inspired, distributed model for the representation of concrete knowledge (verbs, names, attributes) in a system able to recognize simple word compounds and to execute corresponding actions
a multiple constraints satisfaction model able to reproduce linguistic and motor sequences of different levels of complexion

Acknowledgments

This PhD work has been initiated while working with Prof. Ronan Reilly at University College Dublin, Ireland, and has been recently developed under the supervision of Prof. Aude Billard, at Autonomous Systems Lab, EPFL, Suisse. I am grateful to Aude Billard to have given me the chance to work under her supervision at the Autonomous System Lab, EPFL. I am very grateful to Stefan Schaal to have provided access to Xanim simulation environment for the experiments presented here. My research at EPFL has been supported by the Swiss National Foundation of Science.

Relevant papers

Billard, A., and Dautenhahn, K., (2000) Experiments in social robotics: grounding and use of communication in autonomous agents, Adaptive Behavior, vol. 7: 3/4.

Breazeal, C., Hoffman, G., and Lockerd, A., (2004), Teaching and Working with Robots as a Collaboration, submitted to Autonomous Agents and Multi-Agent Systems.

Demiris, J., Hayes, G.M. (2001), Imitation as a dual-route process featuring predictive and learning components: a biologically-plausible computational model, Imitation in Animals and

Artifacts, K. Dautenhahn, C. Nehaniv (Eds.), MIT Press.

Goga, I., Billard, A. (2004), A computational framework for the study of parallel development of manipulatory and linguistic skills. Technical report, Autonomous Systems Laboratory, Swiss Institute of Technology Lausanne (EPFL), Switzerland.

Goga, I., Billard, A. (2006) Attention mechanisms for goal-directed action in cognitive agents. Accepted at the Third Conference Towards Autonomous Robotic Systems TAROS, University of Surrey, UK.

Goga, I., Billard, A. (in press), Development of goal-directed imitation, object manipulation and language in humans and robots. In M. A. Arbib (ed.), Action to Language via the Mirror Neuron System, Cambridge University Press (in press).

Greenfield, P., Nelson, K., and Saltzman, E., 1972, The development of rulebound strategies for manipulating seriated cups: a parallel between action and grammar, Cognitive psychology, 3:291-310.

Schaal, S. (2000). The SL simulation and real-time control software package, USC.

Weng, J., McClelland, J., Pentland, A., Sporns, O., Stockman, I., Sur, M., Thelen, E., (2000) Autonomous mental development by robots and animals, Science, 291:5504, 599 - 600.

Zlatev, J., and Balkenius, C., (2001) Introduction: Why epigenetic robotics?, Proc. First Workshop on Epigenetic Robotics, Lund, Cognitive Studies 85, 2001.