Linking language with embodied and teleological representations of action for humanoid cognition

Events and Causal Mappings Modeled in Conceptual Spaces

The aim of the article is to present a model of causal relations that is based on what is known about human causal reasoning and that forms guidelines for implementations in robots. I argue for two theses concerning human cognition. The first is that human causal cognition, in contrast to that of other animals, is based on the understanding of the forces that are involved. The second thesis is that humans think about causality in terms of events. I present a two-vector model of events, developed by Gärdenfors and Warglien, which states that an event is represented in terms of two main components - the force of an action that drives the event, and the result of its application. Apart from the causal mapping, the event model contains representations of a patient, an agent, and possibly some other roles. Agents and patients are objects (animate or inanimate) that have different properties. Following my theory of conceptual spaces, they can be described as vectors of property values. At least two spaces are needed to describe an event, an action space and a result space. The result of an event is modeled as a vector representing the change of properties of the patient before and after the event. In robotics the focus has been on describing results. The proposed model also includes the causal part of events, typically described as an action. A central part of an event category is the mapping from actions to results. This mapping contains the central information about causal relations. In applications of the two-vector model, the central problem is how the event mapping can be learned in a way that is amenable to implementations in robots. Three processes are central for event cognition: causal thinking, control of action and learning by generalization. Although it is not yet clear which is the best way to model how the mappings can be learned, they should be constrained by three corresponding mathematical properties: monotonicity (related to qualitative causal thinking); continuity (plays a key role in activities of action control); and convexity (facilitates generalization and the categorization of events). I argue that Bayesian models are not suitable for these purposes, but some more geometrically oriented approach to event mappings should be used.

Using Event Representations to Generate Robot Semantics

Most semantic models employed in human-robot interactions concern how a robot can understand commands, but in this article the aim is to present a framework that allows dialogic interaction. The key idea is to use events as the fundamental structures for the semantic representations of a robot. Events are modeled in terms of conceptual spaces and mappings between spaces. It is shown how the semantics of major word classes can be described with the aid of conceptual spaces in a way that is amenable to computer implementations. An event is represented by two vectors, one force vector representing an action and one result vector representing the effect of the action. The two-vector model is then extended by the thematic roles so an event is built up from an agent, an action, a patient, and a result. It is shown how the components of an event can be put together to semantic structures that represent the meanings of sentences. It is argued that a semantic framework based on events can generate a general representational framework for human-robot communication. An implementation of the framework involving communication with an iCub will be described.

Recurrent temporal networks and language acquisition-from corticostriatal neurophysiology to reservoir computing

One of the most paradoxical aspects of human language is that it is so unlike any other form of behavior in the animal world, yet at the same time, it has developed in a species that is not far removed from ancestral species that do not possess language. While aspects of non-human primate and avian interaction clearly constitute communication, this communication appears distinct from the rich, combinatorial and abstract quality of human language. So how does the human primate brain allow for language? In an effort to answer this question, a line of research has been developed that attempts to build a language processing capability based in part on the gross neuroanatomy of the corticostriatal system of the human brain. This paper situates this research program in its historical context, that begins with the primate oculomotor system and sensorimotor sequencing, and passes, via recent advances in reservoir computing to provide insight into the open questions, and possible approaches, for future research that attempts to model language processing. One novel and useful idea from this research is that the overlap of cortical projections onto common regions in the striatum allows for adaptive binding of cortical signals from distinct circuits, under the control of dopamine, which has a strong adaptive advantage. A second idea is that recurrent cortical networks with fixed connections can represent arbitrary sequential and temporal structure, which is the basis of the reservoir computing framework. Finally, bringing these notions together, a relatively simple mechanism can be built for learning the grammatical constructions, as the mappings from surface structure of sentences to their meaning. This research suggests that the components of language that link conceptual structure to grammatical structure may be much simpler that has been proposed in other research programs. It also suggests that part of the residual complexity is in the conceptual system itself.

I Reach Faster When I See You Look: Gaze Effects in Human-Human and Human-Robot Face-to-Face Cooperation

Human-human interaction in natural environments relies on a variety of perceptual cues. Humanoid robots are becoming increasingly refined in their sensorimotor capabilities, and thus should now be able to manipulate and exploit these social cues in cooperation with their human partners. Previous studies have demonstrated that people follow human and robot gaze, and that it can help them to cope with spatially ambiguous language. Our goal is to extend these findings into the domain of action, to determine how human and robot gaze can influence the speed and accuracy of human action. We report on results from a human-human cooperation experiment demonstrating that an agent's vision of her/his partner's gaze can significantly improve that agent's performance in a cooperative task. We then implement a heuristic capability to generate such gaze cues by a humanoid robot that engages in the same cooperative interaction. The subsequent human-robot experiments demonstrate that a human agent can indeed exploit the predictive gaze of their robot partner in a cooperative task. This allows us to render the humanoid robot more human-like in its ability to communicate with humans. The long term objectives of the work are thus to identify social cooperation cues, and to validate their pertinence through implementation in a cooperative robot. The current research provides the robot with the capability to produce appropriate speech and gaze cues in the context of human-robot cooperation tasks. Gaze is manipulated in three conditions: Full gaze (coordinated eye and head), eyes hidden with sunglasses, and head fixed. We demonstrate the pertinence of these cues in terms of statistical measures of action times for humans in the context of a cooperative task, as gaze significantly facilitates cooperation as measured by human response times.

Grounding the Meanings in Sensorimotor Behavior using Reinforcement Learning

The recent outburst of interest in cognitive developmental robotics is fueled by the ambition to propose ecologically plausible mechanisms of how, among other things, a learning agent/robot could ground linguistic meanings in its sensorimotor behavior. Along this stream, we propose a model that allows the simulated iCub robot to learn the meanings of actions (point, touch, and push) oriented toward objects in robot’s peripersonal space. In our experiments, the iCub learns to execute motor actions and comment on them. Architecturally, the model is composed of three neural-network-based modules that are trained in different ways. The first module, a two-layer perceptron, is trained by back-propagation to attend to the target position in the visual scene, given the low-level visual information and the feature-based target information. The second module, having the form of an actor-critic architecture, is the most distinguishing part of our model, and is trained by a continuous version of reinforcement learning to execute actions as sequences, based on a linguistic command. The third module, an echo-state network, is trained to provide the linguistic description of the executed actions. The trained model generalizes well in case of novel action-target combinations with randomized initial arm positions. It can also promptly adapt its behavior if the action/target suddenly changes during motor execution.

Robot cognitive control with a neurophysiologically inspired reinforcement learning model

A major challenge in modern robotics is to liberate robots from controlled industrial settings, and allow them to interact with humans and changing environments in the real-world. The current research attempts to determine if a neurophysiologically motivated model of cortical function in the primate can help to address this challenge. Primates are endowed with cognitive systems that allow them to maximize the feedback from their environment by learning the values of actions in diverse situations and by adjusting their behavioral parameters (i.e., cognitive control) to accommodate unexpected events. In such contexts uncertainty can arise from at least two distinct sources – expected uncertainty resulting from noise during sensory-motor interaction in a known context, and unexpected uncertainty resulting from the changing probabilistic structure of the environment. However, it is not clear how neurophysiological mechanisms of reinforcement learning and cognitive control integrate in the brain to produce efficient behavior. Based on primate neuroanatomy and neurophysiology, we propose a novel computational model for the interaction between lateral prefrontal and anterior cingulate cortex reconciling previous models dedicated to these two functions. We deployed the model in two robots and demonstrate that, based on adaptive regulation of a meta-parameter β that controls the exploration rate, the model can robustly deal with the two kinds of uncertainties in the real-world. In addition the model could reproduce monkey behavioral performance and neurophysiological data in two problem-solving tasks. A last experiment extends this to human–robot interaction with the iCub humanoid, and novel sources of uncertainty corresponding to “cheating” by the human. The combined results provide concrete evidence for the ability of neurophysiologically inspired cognitive systems to control advanced robots in the real-world.