In the context of two related NIH projects supporting scientific collaboration we seek to implement an environment for collaborative information retrieval and analysis based on utility theory.

Clinical research of new recreational drugs and trends requires mining current information from non-traditional text sources. In this work we support such research through the use of multi-dimensional latent text models, such as factorial LDA, that capture orthogonal factors of corpora, creating structured output for researchers to better understand the contents of a corpus. Since a purely unsupervised model is unlikely to discover specific factors of interests to clinical researchers, we modify the structure of factorial LDA to incorporate prior knowledge, including the use of of observed variables, informative priors and background components. The resulting model learns factors that correspond to drug type, delivery method (smoking, injection, etc.), and aspect (chemistry, culture, effects, health, usage). We demonstrate that the improved model yields better quantitative and more interpretable results.

List of organizers of the AAAI Fall Symposium on  Information Retrieval and Knowledge Discovery in Biomedical Text

One common use for swarming agents is in social simulation. This paper reports on such a model developed to track protest activities at the May 2012 NATO summit in Chicago. The use of apoptotic stigmergic agents allows the model to run on-line, consuming two kinds of external data and reporting its results in real time.

Robots are increasingly introduced to work in concert with people in high-intensity domains, such as manufacturing, space exploration and hazardous environments. Although there are numerous studies on human teamwork and coordination in these settings, very little prior work exists on applying these models to human-robot interaction. This paper presents results from ongoing work aimed at translating qualitative methods from human factors engineering into computational models that can be applied to human-robot teaming. We describe a statistical approach to learning patterns of strong and weak agreements in human planning meetings that achieves up to 94% prediction accuracy. We also formulate a human-robot interactive planning method that emulates cross-training, a training strategy widely used in human teams. Results from human subject experiments show statistically significant improvements on team fluency metrics, compared to standard reinforcement learning techniques. Results from these two studies support the approach of modeling and applying common practices in human teaming to achieve more effective and fluent human-robot teaming.

Human control of vehicle swarms faces a dilemma: an operator must be able to exercise precise control over how a mission is executed, but controlling individual vehicles is not scalable. The Proto spatial computing lan- guage offers an intermediate representation, where the motion of a swarm is specified as a vector field, which is then approximated by the movement of individual members (Bachrach, Beal, and McLurkin 2010). I propose that this can be exploited to build a “tactical command” model of swarm control, whereby human “officers” dynamically decompose a swarm into units and task those units to carry out geometric and topological maneuvers under the constraints imposed by the platform. This abstraction may also allow situation awareness interfaces for individual agents to be extended to apply to swarm units.

Knowledge engineering from experimental design (KEfED) is a novel approach based on the dependency relationships that occur between the variables of a scientific study. We used this approach to curate the experimental designs of ten scientific papers from a well-established database of HIV vaccine trials in non-human primates. The KEfED models provide a characteristic, data-oriented signature for each measurement made in the study. We present preliminary analysis of these manually-curated, detailed representations using our own open-source curation tools and show the multi-variate statistical analyses on the resultant models of experimental design. The analyses produced a visualization of the similarities between studies and an account of the dependency relationships across studies. We describe our approach in the context of a knowledge engineering strategy based on creating large-scale domain-independent repositories of experimental observatio

We explore applicability of the situation calculus, the well-known logical framework developed in Artificial Intelligence for representation of dynamic systems, to the task of representing knowledge about processes, actions and events in the natural sciences. In this paper, we concentrate on a case study in the area of organic chemistry. More specifically, we adapt the situation calculus to the task of automating organic synthesis planning on a qualitative level, where the objective is to identify a chain of chemical reactions transforming the given initial molecules into the desired goal molecule. We present two approaches for reasoning about reactions in organic chemistry: a “micro” approach and a “macro” approach. The “micro” approach is a low level approach that explicitly represents the most elementary interactions between molecules during a single chemical reaction, namely the splitting and forming of bonds between atoms. In contrast, the“macro” approach is a higher level approach that treats each chemical reaction (a set of splits and formation of bonds) as an elementary action. Both approaches are implemented in PROLOG. Declarative heuristics are defined to reduce the search space and help the program to find the correct synthesis routes more quickly. We hope that the lessons learned from our successful case study can have discovery potential in other bio-medical sciences. We discuss briefly how the proposed approaches can contribute to solving other research problems and to communicating pathways.

Causality discovery has been one of the core tasks in scientific research since the beginning of human scientific history. In the age of data tsunami, the causality discovery task involves identification of causality among millions of variables which cannot be done manually by humans. However, the identification of causality relationships using artificial intelligence and statistical techniques in non-experimental settings faces several challenges. In this work, we address three of the challenges regarding Granger causality, one of the most popular causality inference techniques. First, we analyze the consistency of two most popular Granger causality techniques and show that the significance test is not consistent in high dimensions. Second, we review our nonparametric generalization of the Lasso-Granger technique called Generalized Lasso Granger (GLG) to uncover Granger causality relationships among irregularly sampled time series. Finally, we describe two techniques to uncover the casual dependence in non-linear datasets. Extensive experiments are provided to show the significant advantages of the proposed algorithms over their state-of-the-art counterparts.

Artificial Intelligence researchers have long sought to understand and replicate processes of scientific discovery. This article discusses Discovery Informatics as an emerging area of research that builds on that tradition and applies principles of intelligent computing and information systems to understand, automate, improve, and innovate processes of scientific discovery.