It has been hypothesized that to maximize learning, intelligent tutoring systems should detect and respond to both cognitive student states, and affective and metacognitive states such as uncertainty. In intelligent tutoring research so far, student state detection is primarily based on information available from a single student-system exchange unit, or turn. However, the features used in the detection of such states may have a temporal component, spanning multiple turns, and may change throughout the tutoring process. To test this hypothesis, an interactive tool was implemented for the visual analysis of prosodic features across a corpus of student turns previously annotated for uncertainty. The tool consists of two complementary visualization modules. The first module allows researchers to visually mine the feature data for patterns per individual student dialogue, and form hypotheses about feature dependencies. The second module allows researchers to quickly test these hypotheses on groups of students through statistical visual analysis of feature dependencies. Results show that significant differences exist among feature patterns across different student groups. Further analysis suggests that feature patterns may vary with student domain knowledge.

The Association for the Advancement of Artificial Intelligence, in cooperation with Stanford University's Department of Computer Science, was pleased to present the 2009 Spring Symposium Series, held Monday through Wednesday, March 23–25, 2009 at Stanford University. The titles of the nine symposia were Agents that Learn from Human Teachers, Benchmarking of Qualitative Spatial and Temporal Reasoning Systems, Experimental Design for Real-World Systems, Human Behavior Modeling, Intelligent Event Processing, Intelligent Narrative Technologies II, Learning by Reading and Learning to Read, Social Semantic Web: Where Web 2.0 Meets Web 3.0, and Technosocial Predictive Analytics. The goal of the Agents that Learn from Human Teachers was to investigate how we can enable software and robotics agents to learn from real-time interaction with an everyday human partner. The aim of the Benchmarking of Qualitative Spatial and Temporal Reasoning Systems symposium was to initiate the development of a problem repository in the field of qualitative spatial and temporal reasoning and identify a graded set of challenges for future midterm and long-term research. The Experimental Design symposium discussed the challenges of evaluating AI systems. The Human Behavior Modeling symposium explored reasoning methods for understanding various aspects of human behavior, especially in the context of designing intelligent systems that interact with humans. The Intelligent Event Processing symposium discussed the need for more AI-based approaches in event processing and defined a kind of research agenda for the field, coined as intelligent complex event processing (iCEP). The Intelligent Narrative Technologies II AAAI symposium discussed innovations, progress, and novel techniques in the research domain. The Learning by Reading and Learning to Read symposium explored two aspects of making natural language texts semantically accessible to, and processable by, machines. The Social Semantic Web symposium focused on the real-world grand challenges in this area. Finally, the Technosocial Predictive Analytics symposium explored new methods for anticipatory analytical thinking that provide decision advantage through the integration of human and physical models.

Despite the fact that AI has always been adventurous in trying to elucidate complex aspects of human behaviour, only recently has there been research into computational modelling of humor. One obstacle to progress is the lack of a precise and detailed theory of how humor operates. Nevertheless, since the early 1990s, there have been a number of small programs that create simple verbal humor, and more recently there have been studies of the automatic classification of the humorous status of texts. In addition, there are a number of advocates of the practical uses of computational humor: in user-interfaces, in education, and in advertising. Computer-generated humor is still quite basic, but it could be viewed as a form of exploratory creativity. For computational humor to improve, some hard problems in AI will have to be addressed.

AAAI-10 will also include several special tracks, including the Nectar and Senior Member tracks, as well as specific research areas. Call for Papers for the main technical track and other tracks are available at www.aaai.org/aaai10.

Many databases store data in relational format, with different types of entities and information about links between the entities. The field of statistical-relational learning (SRL) has developed a number of new statistical models for such data. In this paper we focus on learning class-level or first-order dependencies, which model the general database statistics over attributes of linked objects and links (e.g., the percentage of A grades given in computer science classes). Class-level statistical relationships are important in themselves, and they support applications like policy making, strategic planning, and query optimization. Most current SRL methods find class-level dependencies, but their main task is to support instance-level predictions about the attributes or links of specific entities. We focus only on class-level prediction, and describe algorithms for learning class-level models that are orders of magnitude faster for this task. Our algorithms learn Bayes nets with relational structure, leveraging the efficiency of single-table nonrelational Bayes net learners. An evaluation of our methods on three data sets shows that they are computationally feasible for realistic table sizes, and that the learned structures represent the statistical information in the databases well. After learning compiles the database statistics into a Bayes net, querying these statistics via Bayes net inference is faster than with SQL queries, and does not depend on the size of the database.

We discuss here the mean-field theory for a cellular automata model of meta-learning. The meta-learning is the process of combining outcomes of individual learning procedures in order to determine the final decision with higher accuracy than any single learning method. Our method is constructed from an ensemble of interacting, learning agents, that acquire and process incoming information using various types, or different versions of machine learning algorithms. The abstract learning space, where all agents are located, is constructed here using a fully connected model that couples all agents with random strength values. The cellular automata network simulates the higher level integration of information acquired from the independent learning trials. The final classification of incoming input data is therefore defined as the stationary state of the meta-learning system using simple majority rule, yet the minority clusters that share opposite classification outcome can be observed in the system. Therefore, the probability of selecting proper class for a given input data, can be estimated even without the prior knowledge of its affiliation. The fuzzy logic can be easily introduced into the system, even if learning agents are build from simple binary classification machine learning algorithms by calculating the percentage of agreeing agents.

Explaining policies of Markov Decision Processes (MDPs) is complicated due to their probabilistic and sequential nature. We present a technique to explain policies for factored MDP by populating a set of domain-independent templates. We also present a mechanism to determine a minimal set of templates that, viewed together, completely justify the policy. Our explanations can be generated automatically at run-time with no additional effort required from the MDP designer. We demonstrate our technique using the problems of advising undergraduate students in their course selection and assisting people with dementia in completing the task of handwashing. We also evaluate our explanations for course-advising through a user study involving students.

It has long been recognized that users can have complex preferences on plans.  Non-intrusive learning of such preferences by observing the plans executed by the user is an attractive idea. Unfortunately, the executed plans are often not a true representation of user preferences, as they result from the interaction between user preferences and feasibility constraints. In the travel planning scenario, a user whose true preference is to travel by a plane may well be frequently observed traveling by car because of feasibility constraints (perhaps the user is a poor graduate student). In this work, we describe a novel method for learning true user preferences obfuscated by such feasibility constraints.  Our base learner induces probabilistic hierarchical task networks (pHTNs) from sets of training plans. Our approach is to rescale the input so that it represents the user's preference distribution on plans rather than the observed distribution on plans.

We investigate the issue of model selection and the use of the nonconformity (strangeness) measure in batch learning. Using the nonconformity measure we propose a new training algorithm that helps avoid the need for Cross-Validation or Leave-One-Out model selection strategies. We provide a new generalisation error bound using the notion of nonconformity to upper bound the loss of each test example and show that our proposed approach is comparable to standard model selection methods, but with theoretical guarantees of success and faster convergence. We demonstrate our novel model selection technique using the Support Vector Machine.

We present a streaming model for large-scale classification (in the context of $\ell_2$-SVM) by leveraging connections between learning and computational geometry. The streaming model imposes the constraint that only a single pass over the data is allowed. The $\ell_2$-SVM is known to have an equivalent formulation in terms of the minimum enclosing ball (MEB) problem, and an efficient algorithm based on the idea of \emph{core sets} exists (Core Vector Machine, CVM). CVM learns a $(1+\varepsilon)$-approximate MEB for a set of points and yields an approximate solution to corresponding SVM instance. However CVM works in batch mode requiring multiple passes over the data. This paper presents a single-pass SVM which is based on the minimum enclosing ball of streaming data. We show that the MEB updates for the streaming case can be easily adapted to learn the SVM weight vector in a way similar to using online stochastic gradient updates. Our algorithm performs polylogarithmic computation at each example, and requires very small and constant storage. Experimental results show that, even in such restrictive settings, we can learn efficiently in just one pass and get accuracies comparable to other state-of-the-art SVM solvers (batch and online). We also give an analysis of the algorithm, and discuss some open issues and possible extensions.