Human decisional processes result from the employment of selected quantities of relevant information, generally synthesized from environmental incoming data and stored memories. Their main goal is the production of an appropriate and adaptive response to a cognitive or behavioral task. Different strategies of response production can be adopted, among which haphazard trials, formation of mental schemes and heuristics. In this paper, we propose a model of Boolean neural network that incorporates these strategies by recurring to global optimization strategies during the learning session. The model characterizes as well the passage from an unstructured/chaotic attractor neural network typical of data-driven processes to a faster one, forward-only and representative of schema-driven processes. Moreover, a simplified version of the Iowa Gambling Task (IGT) is introduced in order to test the model. Our results match with experimental data and point out some relevant knowledge coming from psychological domain.

This paper addresses the problem of identifying a lower dimensional space where observed data can be sparsely represented. This under-complete dictionary learning task can be formulated as a blind separation problem of sparse sources linearly mixed with an unknown orthogonal mixing matrix. This issue is formulated in a Bayesian framework. First, the unknown sparse sources are modeled as Bernoulli-Gaussian processes. To promote sparsity, a weighted mixture of an atom at zero and a Gaussian distribution is proposed as prior distribution for the unobserved sources. A non-informative prior distribution defined on an appropriate Stiefel manifold is elected for the mixing matrix. The Bayesian inference on the unknown parameters is conducted using a Markov chain Monte Carlo (MCMC) method. A partially collapsed Gibbs sampler is designed to generate samples asymptotically distributed according to the joint posterior distribution of the unknown model parameters and hyperparameters. These samples are then used to approximate the joint maximum a posteriori estimator of the sources and mixing matrix. Simulations conducted on synthetic data are reported to illustrate the performance of the method for recovering sparse representations. An application to sparse coding on under-complete dictionary is finally investigated.

I introduce an algorithm for estimating parameters from multidimensional data based on forward modelling. In contrast to many machine learning approaches it avoids fitting an inverse model and the problems associated with this. The algorithm makes explicit use of the sensitivities of the data to the parameters, with the goal of better treating parameters which only have a weak impact on the data. The forward modelling approach provides uncertainty (full covariance) estimates in the predicted parameters as well as a goodness-of-fit for observations. I demonstrate the algorithm, ILIUM, with the estimation of stellar astrophysical parameters (APs) from simulations of the low resolution spectrophotometry to be obtained by Gaia. The AP accuracy is competitive with that obtained by a support vector machine. For example, for zero extinction stars covering a wide range of metallicity, surface gravity and temperature, ILIUM can estimate Teff to an accuracy of 0.3% at G=15 and to 4% for (lower signal-to-noise ratio) spectra at G=20. [Fe/H] and logg can be estimated to accuracies of 0.1-0.4dex for stars with G<=18.5. If extinction varies a priori over a wide range (Av=0-10mag), then Teff and Av can be estimated quite accurately (3-4% and 0.1-0.2mag respectively at G=15), but there is a strong and ubiquitous degeneracy in these parameters which limits our ability to estimate either accurately at faint magnitudes. Using the forward model we can map these degeneracies (in advance), and thus provide a complete probability distribution over solutions. (Abridged)

Although AI and HCI explore computing and intelligent behavior and the fields have seen some cross-over, until recently there was not very much. This article outlines a history of the fields that identifies some of the forces that kept the fields at arm’s length. AI was generally marked by a very ambitious, long-term vision requiring expensive systems, although the term was rarely envisioned as being as long as it proved to be, whereas HCI focused more on innovation and improvement of widely-used hardware within a short time-scale. These differences led to different priorities, methods, and assessment approaches.  A consequence was competition for resources, with HCI flourishing in AI winters and moving more slowly when AI was in favor. The situation today is much more promising, in part because of platform convergence: AI can be exploited on widely-used systems.

This is an opinion piece about the relationship between the fields of human-computer interaction (HCI), and artificial intelligence (AI). The ultimate goal of both fields is to make user interfaces more effective and easier to use for people. But historically, they have disagreed about whether "intelligence" or "direct manipulation" is the better route to achieving this. There is an unjustified perception in HCI that AI is unreliable. There is an unjustified perception in AI that interfaces are merely cosmetic. This disagreement is counterproductive.This article argues that AI's goals of intelligent interfaces would benefit enormously by the user-centered design and testing principles of HCI. It argues that HCI's stated goals of meeting the needs of users and interacting in natural ways, would be best served by application of AI. Peace.

We report part of the design experience gained in X-Media, a system for knowledge management and sharing. Consolidated techniques of interaction design (scenario-based design) had to be revisited to capture the richness and complexity of intelligent interactive systems. We show that the design of intelligent systems requires methodologies (faceted scenarios) that support the investigation of intelligent features and usability factors simultaneously. Interaction designers become mediators between intelligent technology and users, and have to facilitate reciprocal understanding.

ICAART 2010 will be held January 22-24, 2010, in Valencia, Spain. This page includes forthcoming AAAI sponsored conferences, conferences presented International Conference on Intelligent by AAAI Affiliates, and conferences held in cooperation with AAAI. IUI 2010 will be Magazine also maintains a calendar listing that includes nonaffiliated conferences held February 7-10, 2010, in Hong at www.aaai.org/Magazine/calendar.php. The Twelfth International Conference The Third Conference on Artificial AAAI Spring Symposium Series will be on Principles of Knowledge Representation General Intelligence. AGI-08 will be held March 22-24, 2010 at Stanford and Reasoning.

General Intelligence, was held March 6-9 in Arlington, Virginia. Pascal Hitzler chaired the program committee. The first day of the conference featured in-depth tutorials on leading AGI systems and approaches, including introductions to the SOAR, Texai, and OpenCog software, and overviews of the logic-based, reinforcement learning and program-induction approaches to AGI. Following this, the main conference on Saturday and Sunday featured a number of themed sessions: Evaluation and Metrics (chaired by John Laird), Robotics and Embodiment (chaired by Itamar Arel), Cognitive Architectures (chaired by Pei Wang and Stephen Reed), Logical Approaches to AGI (chaired by Selmer Bringsjord), Learning and Reasoning (chaired by Selmer Bringsjord), Speech and Language (chaired by Moshe Looks), and Self-Awareness and Consciousness (chaired by Ben Goertzel). There were fewer industry participants because in early 2009 (due to the global economic crisis) many U.S. firms were restricting On the other hand there was an Emanuel Kitzelmann, Martin Hofmann, and Ute even greater international participation, including Schmid, from the Cognitive Systems Group at the a keynote speech by Juergen Schmidhuber (from University of Bamberg, who work in the AI tradition IDSIA, in Lugano, Switzerland, and the Technical of "inductive programing." Their paper University of Munich) and a large number of presentations described a clever way to reformulate the conclusions from German researchers.

The 22nd International Florida Artificial Intelligence Research Society Conference (FLAIRS-22) was held 19th – 21st May 2009 at the Sundial Beach and Golf Resort on Sanibel Island, Florida, USA.  It continued a long tradition of FLAIRS conferences, which attract researchers from around the world.  The conference featured technical papers, special tracks, and invited speakers.  This year’s conference was chaired by Susan Haller, from the State University of New York at Potsdam.  Conference program co-chairs were Hans W. Guesgen, from Massey University in New Zealand, and H. Chad Lane, from the University of Southern California.  The special tracks were coordinated by Philip McCarthy, from the University of Memphis.

The development of intelligent environments is considered an important step toward the realization of the ambient intelligence vision. Greece, served as program chairs. The previous four editions of the IE conference have been held at the University of Essex, UK (in 2005), at the National Technical University of Athens, Greece (in 2006), at the University of Ulm, Germany (in 2007), and at the University of Washington campus in Seattle, Washington, USA (in 2008). The development of intelligent environments is About 120 delegates attended the workshops considered the first and primary step toward the and the conference. These included representatives realization of the ambient intelligence vision.