Electrical Engineering & Computer Science

Grove Karl Gilbert (1843-1918) was one of the most distinguished American geologists. He spent his career entirely in Government service. Gilbert joined the U.S. Geological Survey on its founding in 1879 and served as its first Chief Geologist (1889-1892). This report summarizes Gilbert's life and career, with special attention to his contributions to geomorphology, tectonics, hydrology, glaciology, earthquake studies, and geological methods.

In this paper, we describe new methods for the simulation, measurement and representation of sEMG signals. With regard to simulation, we choose a 2-D state space model and suggest a 3-D model which can account for in-homogeneities, nonlinearities and memory in the medium and which can be hardware accelerated through FPGA. With regard to measurement we use surface electrodes with a new amplifier circuit topology, which mitigates the effects of pickup and artifacts. With regard to representation, we describe a method for using wavelets which shows promise for isolating signals of interest.

Reasoning about social networks (labeled, directed, weighted graphs) is be-
coming increasingly important and there are now models of how certain phenomena (e.g.
adoption of products/services by consumers, spread of a given disease) \diuse" through
the network. Some of these diusion models can be expressed via generalized annotated
programs (GAPs). In this paper, we consider the following problem: suppose we have a
given goal to achieve (e.g. maximize the expected number of adoptees of a product or
minimize the spread of a disease) and suppose we have limited resources to use in trying
to achieve the goal (e.g. give out a few free plans, provide medication to key people in the
SN) - how should these resources be used so that we optimize a given objective function
related to the goal? We dene a class of social network optimization problems (SNOPs)
that supports this type of reasoning. We formalize and study the complexity of SNOPs
and show how they can be used in conjunction with existing economic and disease diusion
models.

There are many applications where we observe various phenomena in space (e.g. locations of victims of a serial killer), and where we want to infer \partner" locations (e.g. the location where the killer lives) that are geospatially related to the observed phenomena. In this paper, we define geospatial abduction problems (GAPs for short). We analyze the complexity of GAPs, develop exact and approximate algorithms (often with approximation guarantees) for these problems together with analyses of these algorithms, and develop a prototype implementation of our GAP framework. We demonstrate accuracy of our algorithms on a real world data set consisting of insurgent IED (improvised explosive device) attacks against US forces in Iraq (the observations were the locations of the attacks, while the \partner" locations we were trying to infer were the locations of IED weapons caches).

The analytical software tools of the future will radically
change intelligence analysis. Currently, the research
community is utilizing the expertise of both
social and computer scientists to produce these
tools. Computer software utilizing artificial intelligence
and geospatial information techniques will leverage
research in the social sciences to conduct
predictive analysis that is not widely performed on
today’s battlefield. Software that takes into account
the behavior of various non-U.S. actors with respect
to terrain will specifically aid in the counterinsurgency
(COIN) fight–particularly at the tactical level.

On June 17th, 2010, security researchers at a small Belarusian firm known as VirusBlockAda identified malicious software (malware) that infected USB memory sticks.1 In the months that followed, there was a flurry of activity in the computer security community – revealing that this discovery identified only one component of a new computer worm2 known as Stuxnet. This software was designed to specifically target industrial equipment. Once it was revealed that the majority of infections were discovered in Iran,3 along with an unexplained decommissioning of centrifuges at the Iranian fuel enrichment plant (FEP) at Natanz,4 many in the media speculated that the ultimate goal of Stuxnet was to target Iranian nuclear facilities. In November of 2010, some of these suspicions were validated when Iranian President Mahmoud Ahmadinejad publically acknowledged that a computer worm created problems for a “limited number of our [nuclear] centrifuges.” 5 Reputable experts in the computer security community have already labeled Stuxnet as “unprecedented,”6 an “evolutionary leap,”7 and “the type of threat we hope to never see again."8
In this paper, I argue that this malicious software represents a revolution in military affairs (RMA)9 in the virtual realm – that is Stuxnet fundamentally changes the nature of cyber warfare. There are four reasons to this claim: (1) Stuxnet represents the first case in which industrial equipment was targeted with a cyber-weapon, (2) there is evidence that the worm was successful in its targeting of such equipment, (3) it represents a significant advance in the development of malicious software, and (4) Stuxnet has shown that several common assumptions about cyber-security are not always valid. In this paper I examine these four points as well as explore the future implications of the Stuxnet RMA.

We designed the common sense analytical framework
introduced in this article to help staff planners
better determine locations on the ground where
caches can be found. In each of our vignettes, the unit
leveraged this information differently. In Northern
Ireland, the British used this analysis to establish a
list of likely indicators for cache sites. These could
be given to patrols looking for suspicious areas or
used to help verify other pieces of intelligence, such
as HUMINT. In Vietnam, good analysis led to a
highly successful operation where munitions were
recovered and additional attacks were prevented by
removing the IED supplies. In Iraq, the analysis led
to an operational decision to deny the enemy use of
the cache locations. The way to use this analysis is
certainly not limited to these techniques. Essentially,
the framework of this paper can be used to aid development
of named areas of interest, which can then be
used in a variety of ways—from kinetic operations
to cueing of intelligence assets.

In this paper we introduce SCARE - the Spatial Cultural
Abductive Reasoning Engine which solves spatial
abduction problems (Shakarian, Subrahmanian, and
Sapino 2009). We review results of SCARE for activities
by Iranian-sponsored “Special Groups” (Kagan,
Kagan, and Pletka 2008) operating throughout the
Baghdad urban area and compare these findings with
new experiments where we predict IED cache sites of
the Special Groups in Sadr City. We find that by localizing
the spatial abduction problem to a smaller area we
obtain greater accuracy - predicting cache sites within
0:33 km as opposed to 0:72 km for all of Baghdad. We
suspect that local factors of physical and cultural geography
impact reasoning with spatial abduction for this
problem.

Geospatial abduction problems (GAPs) involve the inference of a set of locations that "best explain" a given set of locations of observations. For example, the observations might include
locations where a serial killer committed murders or where insurgents carried out Improvised Explosive Device (IED) attacks. In both these cases, we would like to infer a set of locations that explain the observations, e.g., the set of locations where the serial killer lives/works, and the set of locations where insurgents locate weapons caches. However, unlike all past work on abduction, there is a strong adversarial component to this|an adversary actively attempts to prevent us from discovering such locations. We formalize such abduction problems as a two-player game where both players (an "agent" and an "adversary") use a probabilistic model of their opponent (i.e., a mixed strategy). There is asymmetry as the adversary can choose both the locations of the observations and the locations of the explanation, while the agent (i.e., us) tries to discover these. In this paper, we study the problem from the point of view of both players. We define reward functions axiomatically to capture the similarity between two sets of explanations (one corresponding to the locations chosen by the adversary, one guessed by the agent). Many different reward functions can satisfy our axioms. We then formalize the optimal adversary strategy (OAS) problem and the maximal counter-adversary strategy (MCA) and show that both are NP-hard, that their associated counting complexity problems are #P-hard, and that MCA has no fully polynomial approximation scheme unless P=NP. We show that approximation guarantees are possible for MCA when the reward function satisfies two simple properties (zero-starting and monotonicity) which many natural reward functions satisfy. We develop a mixed integer linear programming algorithm to solve OAS and two algorithms to (approximately) compute MCA; the algorithms yield different approximation guarantees and one algorithm assumes a monotonic reward function. Our experiments use real data about IED attacks over a 21 month period in Baghdad. We are able to show that both the MCA algorithms work well in practice; while MCA-GREEDY-MONO is both highly accurate and slightly faster than MCA-LS, MCA-LS (to our surprise) always completely and correctly maximized the expected benefit to the agent while running in an acceptable time period.

The mobile workforce is rapidly increasing, and technological advances make it feasible for these workers to have ubiquitous access to a variety of resources with various protection requirements. The dynamic computing environment of these workers mandates a security paradigm in which security is tightly coupled with a user's current situation. We have proposed a security paradigm to achieve this, called Shrink-Wrapped Security, in which security is constantly adapting to a user's current situation, and a comprehensive amount of security-relevant context is used to characterize a user's situation. We present an approach that uses generalized annotated programs (GAPs) to practically incorporate such context into security services, with a focus on access control. This allows us to represent context in a principled manner; consistently make security-related decisions; easily make temporary, ad-hoc changes to a security policy; and give a user feedback when access is denied so that she can make the appropriate adjustments.

In evolutionary graph theory biologists study the problem of determining the probability that a small number of mutants overtake a population that is structured on a weighted, possibly directed graph. Currently Monte Carlo simulations are used for estimating such fixation probabilities on directed graphs, since no good analytical methods exist. In this paper, we introduce a novel deterministic algorithm for computing fixation probabilities for strongly connected directed, weighted evolutionary graphs under the case of neutral drift, which we show to be a lower bound for the case where the mutant is more fit than the rest of the population (previously, this was only observed from simulation). We also show that, in neutral drift, fixation probability is additive under the weighted, directed case. We implement our algorithm and show experimentally that it consistently outperforms Monte Carlo simulations by several orders of magnitude, which can allow researchers to study fixation probability on much larger graphs.

Imagine yourself as a military officer in a conflict zone trying to identify locations of weapons caches supporting road-side bomb attacks on your country’s troops. Or imagine yourself as a public health expert trying to identify the location of contaminated water that is causing diarrheal diseases in a local population. Geospatial abduction is a new technique introduced by the authors that allows such problems to be solved. Geospatial Abduction provides the mathematics underlying geospatial abduction and the algorithms to solve them in practice; it has wide applicability and can be used by practitioners and researchers in many different fields. Real-world applications of geospatial abduction to military problems are included. Compelling examples drawn from other domains as diverse as criminology, epidemiology and archaeology are covered as well. This book also includes access to a dedicated website on geospatial abduction hosted by University of Maryland. Geospatial Abduction targets practitioners working in general AI, game theory, linear programming, data mining, machine learning, and more. Those working in the fields of computer science, mathematics, geoinformation, geological and biological science will also find this book valuable.

In August 2008, the Russian Army invaded georgia. numerous, coordinated cyber attacks accompanied the military campaign. this represents the first instance of a large-scale computer network attack (CNA) conducted in tandem with major ground combat operations. The attack had no direct connection to the Russian government, but had a significant informational and psychological impact on Georgia: it effectively isolated the Caucasus state from the outside world.

In a "tipping" model, each node in a social network, representing an individual, adopts a behavior if a certain number of his incoming neighbors previously held that property.  A key problem for viral marketers is to determine an initial "seed" set in a network such that if given a property then the entire network adopts the behavior.  Here we introduce a method for quickly finding seed sets that scales to very large networks.  Our approach finds a set of nodes that guarantees spreading to the entire network under the tipping model.  After experimentally evaluating 31 real-world networks, we found that our approach often finds such sets that are several orders of magnitude smaller than the population size.  Our approach also scales well - on a Friendster social network consisting of 5.6 million nodes and 28 million edges we found a seed sets in under 3.6 hours.  We also find that highly clustered local neighborhoods and dense network-wide community structure together suppress the ability of a trend to spread under the tipping model.