Sven Brueckner’s research while affiliated with Jacobs Engineering Group Inc and other places

Self-organizing software systems are an increasingly attractive approach to highly distributed, decentralized, dynamic applications. In some domains (such as the Internet), the interaction of originally independent systems yields a self-organizing system de facto , and engineers must take these characteristics into account to manage them. This review surveys current work in this field and outlines its main themes, identifies challenges for future research, and addresses the continuity between software engineering in general and techniques appropriate for self-organizing systems.

In most information retrieval systems, software processes (whether agent-based or not) reason about passive items of data. An alternative approach instantiates each record as an agent that actively self-organizes with other agents (including queries). Imitating the movement of bodies under physical forces, we describe a distributed algorithm (“force-based clustering,” or FBC) for dynamically clustering and querying large, heterogeneous, dynamic collections of entities. The algorithm moves entities in a virtual space in a way that estimates the transitive closure of the pairwise comparisons. We demonstrate this algorithm on a large, heterogeneous collection of records, each representing a person. We have some information about a person of interest, but no record in the collection directly matches this information. Application of FBC identifies a small subset of records that are good candidates for describing the person of interest, for further manual investigation and verification.

Computational Social Science (CSS) models are most commonly used to articulate theories and explore their implications. As they become more mature, they are also valuable in monitoring real-world situations. Such applications require models to be linked to dynamic real-world data in real time. This paper explores this distinction in a specific application that tracks crowd violence in an urban setting.

In many applications, agents (whether human or computational) provide estimates that must be combined at a higher level. Recent research distinguishes two kinds of such estimates: interpreted and generated data. These two kinds of data require different kinds of aggregation processes, which behave differently from an information geometric perspective: interpreted estimates require methods such as voting that can leave the convex hull of the individual estimates, while the optimal aggregation for generated estimates lies within the convex hull and thus is accessible by methods such as weighted averages. We motivate our analysis in the context of a crowdsourced forecasting application, demonstrate the central insights theoretically, and show how these insights manifest them-selves in actual data.

In many contexts, people generate forecasts about events of interest, and decision-makers wish to aggregate these forecasts to improve their accuracy. These forecasts differ from signals in the physical sciences. In particular, sensor signals are noisy samples from a common underlying distribution, while human-generated forecasts are based on cognitive models that vary from one informant to another. As a result, human forecasts, unlike physical signals, are not guaranteed to be statisticallyindependent conditioned on the true outcome. These differences both provide new opportunities for aggregation, and impose restrictions that do not apply to physical signals. This paper describes the difference between forecasts and physical signals, outlines a strategy for exploiting these differences in aggregation, and demonstrates modest but statistically significant gains in the accuracy of aggregated forecasts using data from a large ongoing experiment in forecasting world events.

Researchers have explored many formalisms to model how people think about their world. We describe an application that requires modeling how people forecast events in the real world. The naïve assumption is that they use formalisms that model how the world actually evolves. These formalisms are at variance with empirical psychological results. We present a more realistic alternative, the Narrative Space Model (NSM), describe a swarming agent algorithm to fit its parameters from observed data, and present some early results.

Sappelsa, L., Parunak, H.V.D. and Brueckner, S. (2013). The Generic Narrative Space Model as an Intelligence Analysis Tool. American Intelligence Journal, 31(2), 69-78. Intelligence analysts have long found profit in graphical representations of their mental models. The discipline of constructing such a model can sharpen and refine the analyst’s understanding, and externalizing the model enables colleagues to review and augment it. We report on a specialization of the concept map, the Narrative Space Model (NSM), which captures concisely the set of coherent narratives that explain how the world may evolve. We define the notion of a narrative space, compare it with other representational frameworks, discuss the thought processes involved in its construction, and within the context of a current Intelligence Advanced Research Project Activity (IARPA) research program, present a narrative space for a specific intelligence question, and a generic domain narrative space for a class of related questions.

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. Copyright © 2012, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.

At a tactical level, insurgents planning attacks with Improvised Explosive Devices (IEDs) are constrained in their choice of target by the specific location of their safe house or weapons cache, the geographic context in which they operate, and the pattern of potential targets as it presents itself at a given time. Geographic profiling in law-enforcement already takes advantage of similar constraints to identify possible origin locations of serial offenders. We show how geographic profiling of past IED events can significantly enhance our ability to identify areas at risk for future attacks. Specifically, we introduce three tightly coupled swarming pattern analysis models (profiling, clustering, forecasting) that refine each others' conclusions dynamically and point to systematic evaluation experiments that confirm the research hypothesis.