Mobile robots are an increasingly important part of search and rescue efforts as well as military combat. 
In order for users to accept these robots and use them effectively, the user must be able to communicate clearly with the robots and obtain explanations of the robots' behavior that will allow the user to understand its actions. 
This paper describes part of a system of software that will be able to produce explanations of the robots' behavior and situation in an interaction with a human operator.

TeamTalk is dialogue framework that supports multi-participant spoken interaction between humans and robots in a task-oriented setting that requires cooperation and coordination between team members. This paper describes some recently added features to the system, in particular the ability for robots to accept and remember location labels and the ability to learn action sequences. These capabilities reflect the incorporation into the system of an ontology and an instruction understanding component.

Quantum theory has recently been employed to further advance thetheory of information retrieval (IR). A challenging research topicis to investigate the so called quantum-like interference in users'relevance judgment process, where users are involved to judge therelevance degree of each document with respect to a given query. Inthis process, users' relevance judgment for the current document isoften interfered by the judgment for previous documents, due to theinterference on users' cognitive status. Research from cognitivescience has demonstrated some initial evidence of quantum-likecognitive interference in human decision making, which underpins theuser's relevance judgment process. This motivates us to model suchcognitive interference in the relevance judgment process, which inour belief will lead to a better modeling and explanation of userbehaviors in relevance judgement process for IR and eventually leadto more user-centric IR models. In this paper, we propose to useprobabilistic automaton (PA) and quantum finite automaton (QFA),which are suitable to represent the transition of user judgmentstates, to dynamically model the cognitive interference when theuser is judging a list of documents.

Developing the capabilities of children to comprehend written texts is key to their development as young adults. Text comprehension skills develop enormously from the age of 7- 8 until the age of 11. Nowadays, several young children (˜5% – 10% of novice readers) turn out to be poor (text) comprehenders: they demonstrate text comprehension difficulties, related to inference-making skills, despite proficiency in lowlevel cognitive skills like word decoding. Though there are several pencil-and-paper reading interventions for improving inference-making skills on text, and addressed to poor comprehenders, the design and evaluation of Adaptive Learning Systems (ALSs) are lagging behind. The use of more intelligent ALSs to custom-tailor such interventions in the form of games for poor comprehenders has tremendous potential. Our system embodies that potential. This paper presents the design of our ALS by focusing on its intelligent adaptive engine and the related conceptual models, and by presenting the visual interfaces for story telling and gaming.

In this paper we investigate the performance of Cultural Algorithms over the complete range of system complexities, from fixed to chaotic.In order to apply the Cultural Algorithm over all complexity classes we generalize on its co-evolutionary nature to keep the variation in the population across all complexities. Based on previous cultural algorithm approaches, we were to extend the existing models to produce a more general one that could be applied across all complexity classes. We produced a new version of the Cultural Algorithms Toolkit, CAT 2.0, which supported a variety of co-evolutionary features at both the Knowledge and Population levels. We then applied the system to the solution of a 150 randomly generated problems that ranged from simple to chaotic complexity classes. As a result we were able to produce the following conclusions: No homogeneous Social Fabric tested was dominant over all categories of complexity. As the complexity of problems increased, so did the complexity of the Social Fabric that was need to deal with it efficiently. In other words, there was experimental evidence that social structure can be related to the frequency and complexity type of the problems that presented to a cultural system.

The ability to take intelligent actions in real-world domains is a goal of great interest in the machine learning community. Unfortunately, the real-world is filled with systems that can bepartially observed but cannot, as yet, be described by first principlemodels. Moreover, the traditional paradigm of direct interaction with the environment used in reinforcement learning (RL) is often prohibitively expensive in practice. An alternative approach that simultaneously solves both of these problems is to gain experience in simulation; the simulation in this approach is a computational model derived from observations. Advances in sensory and information technology are simplifying the acquisition and distribution of real-world datasets to computational scientists; thus, the barrier to linking intelligent control with real-world domains is becoming one of identifying high-quality state-space and transition functions directly from observations. From a dynamical systems perspective, this barrier is analogous to the problem of finding high-quality manifold embeddings and a rich literature of theory and practice exists to address it. The contribution of this work is two-fold. First, we describe an approach for learning optimal control strategies directly from observations using manifold embeddings as the intermediate state representation. Second, we demonstrate how control strategies constructed in this way can answer important scientific questions. As a concrete example, we use our approach to guide experimental decisions in neurostimulation treatments of epilepsy.

What commonsense knowledge do intelligent systems need, in order to recover from failures or deal with unexpected situations? It is impractical to represent predetermined solutions to deal with every unanticipated situation or provide predetermined fixes for all the different ways in which systems may fail. We contend that intelligent systems require only a finite set of anomaly-handling strategies to muddle through anomalous situations. We describe a generalized metacognition module that implements such a set of anomaly-handling strategies and that in principle can be attached to any host system to improve the robustness of that system. Several implemented studies are reported, that support our contention.

In this paper we make a simple theoretical point using a practical issue as an example. The simple theoretical point is that robustness is not 'all or nothing': in asking whether a system is robust one has to ask 'robust with respect to what property?' and 'robust over what set of changes in the system?' The practical issue used to illustrate the point is an examination of degrees of linkage between sub-networks and a pointed contrast in robustness and fragility between the dynamics of (1) contact infection and (2) information transfer or belief change. Time to infection across linked sub-networks, it turns out, is fairly robust with regard to the degree of linkage between them. Time to infection is fragile and sensitive, however, with regard to the type of sub-network involved: total, ring, small world, random, or scale-free. Aspects of robustness and fragility are reversed where it is belief updating with reinforcement rather than infection that is at issue. In information dynamics, the pattern of time to consensus is robust across changes in network type but remarkably fragile with respect to degree of linkage between sub-networks. These results have important implications for public health interventions in realistic social networks, particularly with an eye to ethnic and socio-economic sub-communities, and in social networks with sub-communities changing in structure or linkage.

Attitudes play a significant role in determining how individuals process information and behave. In this paper we have developed a new computational model of population wide attitude change that captures the social level: how individuals interact and communicate information, and the cognitive level: how attitudes and concept interact with each other. The model captures the cognitive aspect by representing each individuals as a parallel constraint satisfaction network. The dynamics of this model are explored through a simple attitude change experiment where we vary the social network and distribution of attitudes in a population.

Games with A Purpose (GWAP) has been demonstrated to be efficient in collecting large amount of knowledge from online users, e.g. Verbosity and Virtual Pet game. However, its effectiveness in knowledge base (KB) construction has not been explored in previous research. This paper examines the knowledge collected in the Vir- tual Pet game and presents an approach to collect more knowledge driven by the existing relations in KB. In this paper, goal-oriented knowledge collection successfully draws 10572 answers for the "food” domain. The answers are verified by online voting to show that 92.07% of them are good sentences and 95.89% of them are new sentences. This result is a significant improvement over the original Virtual Pet game, with 80.58% good sentences and 67.56% weekly new information.