This thesis describes EM-ONE, an architecture for commonsense thinking capable of reflective reasoning about situations involving physical, social, and mental dimensions. EM-ONE uses as its knowledge base a library of commonsense narratives, each describing the physical, social, and mental activity that occurs during an interaction between several actors. EM-ONE reasons with these narratives by applying "mental critics, " procedures that debug problems that exist in the outside world or within EM-ONE itself. Mental critics draw upon commonsense narratives to suggest courses of action, methods for deliberating about the circumstances and consequences of those actions, and—when things go wrong—ways to reflect upon and debug the activity of previously invoked mental critics. Mental critics are arranged into six layers, the reactive, deliberative, reflective, self-reflective, self-conscious, and self-ideals layers. The selection of mental critics within these six layers is itself guided by a separate collection

Abstract not found

Research on algorithms and representations once dominated AI. Recently the importance of architectures has been acknowledged, but researchers have different objectives, presuppositions and conceptual frameworks, and this, can lead to confused terminology, argumentation at cross purposes, re-invention of wheels and fragmentation of the research. We propose a methodological framework: develop a representation of a general class of architectures within which different architectures can be compared and contrasted. This should facilitate communication and integration across sub-fields of and approaches to AI, as well providing a framework for evaluating alternative architectures. As a first-draft example we present the CogAff architecture schema, and show how it provides a draft framework. But there is much still to be done.

■ To build a machine that has “common sense ” was once a principal goal in the field of artificial intelligence. But most researchers in recent years have retreated from that ambitious aim. Instead, each developed some special technique that could deal with some class of problem well, but does poorly at almost everything else. We are convinced, however, that no one such method will ever turn out to be “best, ” and that instead, the powerful AI systems of the future will use a diverse array of resources that, together, will deal with a great range of problems. To build a machine that’s resourceful enough to have humanlike common sense, we must develop ways to combine the advantages of multiple methods to represent knowledge, multiple ways to make inferences, and multiple ways to learn. We held a two-day symposium in St. Thomas, U.S. Virgin Islands, to discuss such a project—to develop new architectural schemes that can bridge between different strategies and representations. This article reports on the events and ideas developed at this meeting and subsequent thoughts by the authors on how to make progress.

Animals and robots perceiving and acting in a world require an ontology that accommodates entities, processes, states of a#airs, etc., in their environment. If the perceived environment includes information-processing systems, the ontology should reflect that. Scientists studying such systems need an ontology that includes the first-order ontology characterising physical phenomena, the second-order ontology characterising perceivers of physical phenomena, and a (recursive) third order ontology characterising perceivers of perceivers, including introspectors. We argue that second- and third-order ontologies refer to contents of virtual machines and examine requirements for scientific investigation of combined virtual and physical machines, such as animals and robots. We show how the CogA# architecture schema, combining reactive, deliberative, and meta-management categories, provides a first draft schematic third-order ontology for describing a wide range of natural and artificial agents. Many previously proposed architectures use only a subset of CogA#, including subsumption architectures, contention-scheduling systems, architectures with `executive functions' and a variety of types of `Omega' architectures.

This paper presents a critical view of the status quo of some of the emerging research efforts on emotional agents. It attempts to isolate (at least roughly) some of the reasons why emotional agents may be desirable and points to the difficulties of making notions of emotion precise. It lists various problems connected to emotional agents and concludes that it is counterproductive to the whole endeavor of understanding and modeling emotions if "emotion labels" are conferred upon states of agents prematurely without justification.

Abstract. A child, or young human-like robot of the future, needs to develop an information-processing architecture, forms of representation, and mechanisms to support perceiving, manipulating, and thinking about the world, especially perceiving and thinking about actual and possible structures and processes in a 3-D environment. The mechanisms for extending those representations and mechanisms, are also the core mechanisms required for developing mathematical competences, especially geometric and topological reasoning competences. Understanding both the natural processes and the requirements for future human-like robots requires AI designers to develop new forms of representation and mechanisms for geometric and topological reasoning to explain a child’s (or robot’s) development of understanding of affordances, and the proto-affordances that underlie them. A suitable multi-functional self-extending architecture will enable those competences to be developed. Within such a machine, human-like mathematical learning will be possible. It is argued that this can support Kant’s philosophy of mathematics, as against Humean philosophies. It also exposes serious limitations in studies of mathematical development by psychologists. Keywords: learning mathematics, philosophy of mathematics, robot 3-D vision, self-extending architecture, epigenetic robotics 1 Introduction: Approaches

Research on algorithms and representations once dominated AI. Recently the importance of architectures has been acknowledged, but researchers have different objectives, presuppositions and conceptual frameworks, and this can lead to confused terminology, argumentation at cross purposes, re-invention of wheels and fragmentation of the research. We propose a methodological framework: develop a general representation of a wide class of architectures within which different architectures can be compared and contrasted. This should facilitate communication and integration across sub-fields of and approaches to AI, as well as providing a framework for evaluating alternative architectures. As a first-draft example we present the CogAff architecture schema, and show how it provides a useful framework for comparing and contrasting a wide range of architectures, including H-Cogaff, a proposed architecture for human-like systems. All of these concern virtual machine architectures whose natural implementations use biological mechanisms but some of which may use products of human engineering. Besides attempting to understand what sorts of virtual machine architectures are possible

This paper presents a Dynamic Emotion Representation (DER) model, its implementation and an instance of a full humanoid emotional model built with it. The DER model has been implemented to enable users to create their own DER. The instance of the DER model described in this paper represents three types of emotions, primary, secondary emotions and mood. The design of this DER is discussed with reference to emotion theories and to the needs of a facial animation system in which it is integrated. The DER is used in our Emotionally Expressive Facial Animation System (EE-FAS) to produce emotional expressions, to select facial signals corresponding communicative functions in relation to the emotional state of the agent and also in relation to the comparison between the emotional state and the intended meanings expressed through communicative functions.

Abstract. This paper reports experiments demonstrating that the extent to which subjects ascribe emotions to VR faces is highly dependent on textures applied to the face. We demonstrate this for both a photo-realistic vs. non-photo-realistic texture pair and for a male vs. female texture pair. In both cases, experiments were conducted over the Internet on still frames taken from a well-controlled VR emotion modelling system. Given the enormous extent to which textures determine emotion recognition, we consider this a critical area for future research in affective virtual agents. 1