The dynamical and uncertain environments of domestic service robots, which include humans, require rethinking of the benchmarking principles for testing these robots. In RoboCup@Home, statistical procedures are used to track and steer the progress of domestic service robots since 2006. This paper explains the procedures and shows outcomes of these international benchmarking efforts. Although aspects such as shopping in a supermarket receive a fair amount of attention in the robotics community, the authors think that a recently started test is the most important outcome of RoboCup@Home, namely the benchmarking of robot cognition.

We propose a novel method of self-tracking cognitive and neuroanatomical data for monitoring and manipu- lating brain structure. Neuroprofiling, a web service for personalized brain data visualization, is a platform for collecting, analyzing and displaying neuroimaging and cognitive data in a meaningful, simple and individualized way. Users’ T1-weighted, volumetric magnetic resonance imaging (MRI) scans of their brain can be uploaded onto a website, which automatically segments, preprocesses and analyses the image using voxel-based morphometry (VBM) to produce a z-score statistical map of the brain. This reveals, for each subject, the cortical density of different brain regions relative to the mean densities of the population sample. Through a 3D interactive brain-model interface, users can review their unique neuroanatomical profile. By completing behavioral cognitive testing, they can assess different strengths and weaknesses in cognitive abilities that relate to their brain physiology. They can also compare brains with all other users as a group, as well as with experts across different skill domains.

A significant shift is underway as the fields of health and biology are re-organizing into the larger ecosystems of information sciences and complexity sciences. The era of big data is transforming all economic sectors including health and biology. Three big health data streams are being integrated into a standardized investigative method in the realization of personalized medicine – creating individualized risk profiles and interventions such that medical conditions may be combatted during the 80% of their life-cycle while they are still pre-clinical. These three big health data streams are traditional medical data, ‘omics’ data (genomics, microbiomics, proteomics, etc.), and biometric quantified-self daily analytic data. Sequencing costs have continued to decrease such that consumer ‘omics’ data is increasingly available. Simultaneously, the potentially fast-arriving wearable electronics platform (smartwatches, disposable patches, augmented eyewear, etc.) means that it could become possible to unobtrusively collect vast amounts of previously-unavailable objective metric data for each individual and parlay this into personalized physical and mental health optimization platforms. Two experimental protocols are presented here putting this model of integrated health data streams into action and extending recent social intelligence genomics research into the realm of cognitive performance genomics. The DIYgenomics Quantified Creativity study investigates potential linkage between personal genomics and the creative process of the individual. The DIYgenomics Thinking Fast and Slow study examines cognitive bias in thinking (loss aversion and optimism bias) versus personal genomic profiles. The studies integrate big health data streams including traditional health data, personal genomics, quantified self-reported data, standardized questionnaires, and personalized intervention.

Understanding the human mind and increasing individual happiness are important goals in artificial intelligence (AI) and well-being science. The recent revolution in portable self-tracking devices in the data-driven wellness movement and participatory-driven wellness communities, such as the Quantified Self community, provides us with new opportunities to collect psychological or physiological data for understanding the human mind.  While new technologies make it possible to track our daily behavior and various biological signals such as physiological or genetic data more easily, one of the important remaining challenges is to discover our own truly meaningful personal values. Citizen science, scientific research by crowdsourcing or human-based computation, is a new and challenging framework that promotes interdisciplinary research in the fields of computer science, life/brain science, and social psychological/behavioral science, which may introduce new paradigms to the AI community. We have been working on citizen science projects related to the area of personal genomics and have developed a personal genomics information environment named MyFinder. The developed platform supports the search for our inherited talents and maximizes our potential for a meaningful life. In particular, we are interested in the human mind and the personal genome. In this paper, we introduce our MyFinder Project and present the results of a recent study on “social intelligence genomics and empathy building”, and discuss issues involved in exploring our mind within the context of personal genomics.

Belief networks are a new, potentially important, class of knowledge-based models. ARCO1, currently under development at the Atlantic Richfield Company (ARCO) and the University of Southern California (USC), is the most advanced reported implementation of these models in a financial forecasting setting. ARCO1's underlying belief network models the variables believed to have an impact on the crude oil market. A pictorial market model-developed on a MAC II- facilitates consensus among the members of the forecasting team. The system forecasts crude oil prices via Monte Carlo analyses of the network. Several different models of the oil market have been developed; the system's ability to be updated quickly highlights its flexibility.

This paper resolves a common complexity issue in the Bethe approximation of statistical physics and the Belief Propagation (BP) algorithm of artificial intelligence. The Bethe approximation and the BP algorithm are heuristic methods for estimating the partition function and marginal probabilities in graphical models, respectively. The computational complexity of the Bethe approximation is decided by the number of operations required to solve a set of non-linear equations, the so-called Bethe equation. Although the BP algorithm was inspired and developed independently, Yedidia, Freeman and Weiss (2004) showed that the BP algorithm solves the Bethe equation if it converges (however, it often does not). This naturally motivates the following question to understand limitations and empirical successes of the Bethe and BP methods: is the Bethe equation computationally easy to solve? We present a message-passing algorithm solving the Bethe equation in a polynomial number of operations for general binary graphical models of n variables where the maximum degree in the underlying graph is O(log n). Our algorithm can be used as an alternative to BP fixing its convergence issue and is the first fully polynomial-time approximation scheme for the BP fixed-point computation in such a large class of graphical models, while the approximate fixed-point computation is known to be (PPAD-)hard in general. We believe that our technique is of broader interest to understand the computational complexity of the cavity method in statistical physics.

The relationship between belief networks and relational databases is examined. Based on this analysis, a method to construct belief networks automatically from statistical relational data is proposed. A comparison between our method and other methods shows that our method has several advantages when generalization or prediction is deeded.

In mechanical design, there is often unavoidable uncertainty in estimates of design performance. Evaluation of design alternatives requires consideration of the impact of this uncertainty. Expert heuristics embody assumptions regarding the designer's attitude towards risk and uncertainty that might be reasonable in most cases but inaccurate in others. We present a technique to allow designers to incorporate their own unique attitude towards uncertainty as opposed to those assumed by the domain expert's rules. The general approach is to eliminate aspects of heuristic rules which directly or indirectly include assumptions regarding the user's attitude towards risk, and replace them with explicit, user-specified probabilistic multi attribute utility and probability distribution functions. We illustrate the method in a system for material selection for automobile bumpers.

Survey of several forms of updating, with a practical illustrative example. We study several updating (conditioning) schemes that emerge naturally from a common scenarion to provide some insights into their meaning. Updating is a subtle operation and there is no single method, no single 'good' rule. The choice of the appropriate rule must always be given due consideration. Planchet (1989) presents a mathematical survey of many rules. We focus on the practical meaning of these rules. After summarizing the several rules for conditioning, we present an illustrative example in which the various forms of conditioning can be explained.

This paper proposes a new method for solving Bayesian decision problems. The method consists of representing a Bayesian decision problem as a valuation-based system and applying a fusion algorithm for solving it. The fusion algorithm is a hybrid of local computational methods for computation of marginals of joint probability distributions and the local computational methods for discrete optimization problems.