Laura E. Brown

The max-min hill-climbing Bayesian network structure learning algorithm

We present a new algorithm for Bayesian network structure learning, called Max-Min Hill-Climbing (MMHC). The algorithm combines ideas from local learning, constraint-based, and search-and-score techniques in a principled and effective way. It first reconstructs the skeleton of a Bayesian network and then performs a Bayesian-scoring greedy hill-climbing search to orient the edges. In our extensive empirical evaluation MMHC outperforms on average and in terms of various metrics several prototypical and state-of-the-art algorithms, namely the PC, Sparse Candidate, Three Phase Dependency Analysis, Optimal Reinsertion, Greedy Equivalence Search, and Greedy Search. These are the first empirical results simultaneously comparing most of the major Bayesian network algorithms against each other. MMHC offers certain theoretical advantages, specifically over the Sparse Candidate algorithm, corroborated by our experiments. MMHC and detailed results of our study are publicly available at http://www.dsl-lab.org/supplements/mmhc_paper/mmhc_index.html.

Autominder: an intelligent cognitive orthotic system for people with memory impairment

The world’s population is aging at a phenomenal rate. Certain types of cognitive decline, in particular some forms of memory impairment, occur much more frequently in the elderly. This paper describes Autominder, a cognitive orthotic system intended to help older adults adapt to cognitive decline and continue the satisfactory performance of routine activities, thereby potentially enabling them to remain in their own homes longer. Autominder achieves this goal by providing adaptive, personalized reminders of (basic, instrumental, and extended) activities of daily living. Cognitive orthotic systems on the market today mainly provide alarms for prescribed activities at fixed times that are specified in advance. In contrast, Autominder uses a range of AI techniques to model an individual’s daily plans, observe and reason about the execution of those plans, and make decisions about whether and when it is most appropriate to issue reminders. Autominder is currently deployed on a mobile robot, and is being developed as part of the Initiative on Personal Robotic Assistants for the Elderly (the Nursebot project).

Survey of multi-agent systems for microgrid control

Multi-agent systems (MAS) consist of multiple intelligent agents that interact to solve problems that may be beyond the capabilities of a single agent or system. For many years, conceptual MAS designs and architectures have been proposed for applications in power systems and power engineering. With the increasing use and modeling of distributed energy resources for microgrid applications, MAS are well suited to manage the size and complexity of these energy systems. The purpose of this paper is to survey applications of MAS in the control and operation of microgrids. The paper will review MAS concepts, architectures, develop platforms and processes, provide example applications, and discuss limitations.

A novel algorithm for scalable and accurate Bayesian network learning

Bayesian Networks (BN) is a knowledge representation formalism that has been proven to be valuable in biomedicine for constructing decision support systems and for generating causal hypotheses from data. Given the emergence of datasets in medicine and biology with thousands of variables and that current algorithms do not scale more than a few hundred variables in practical domains, new efficient and accurate algorithms are needed to learn high quality BNs from data. We present a new algorithm called Max-Min Hill-Climbing (MMHC) that builds upon and improves the Sparse Candidate (SC) algorithm; a state-of-the-art algorithm that scales up to datasets involving hundreds of variables provided the generating networks are sparse. Compared to the SC, on a number of datasets from medicine and biology, (a) MMHC discovers BNs that are structurally closer to the data-generating BN, (b) the discovered networks are more probable given the data, (c) MMHC is computationally more efficient and scalable than SC, and (d) the generating networks are not required to be uniformly sparse nor is the user of MMHC required to guess correctly the network connectivity

Detection of coherent groups of generators and the need for system separation using synchrophasor data

Detection of groups of coherent generators is very important for balanced and optimal power system separation. Synchrophasor technology has made it possible to get the real time phasors i-e voltage and current along with the frequency. This real time information can be used to find groups of coherent generators. Hierarchical clustering algorithm is used on the rate of change of the generator bus voltage phase angles and rate of change of mean of the generator bus voltage phase angles are tested for the purpose of detection of groups of coherent generators. Initial results show that these approaches can be used for the purpose of real time detection of the groups of coherent generators.

Modeling Cross-Architecture Co-Tenancy Performance Interference

Cloud computing has become a dominant computing paradigm to provide elastic, affordable computing resources to end users. Due to the increased computing power of modern machines powered by multi/many-core computing, data centers often co-locate multiple virtual machines (VMs) into one physical machine, resulting in co-tenancy, and resource sharing and competition. Applications or VMs co-locating in one physical machine can interfere with each other despite of the promise of performance isolation through virtualization. Modelling and predicting co-run interference therefore becomes critical for data center job scheduling and QoS (Quality of Service) assurance. Co-run interference can be categorized into two metrics, sensitivity and pressure, where the former denotes how an applications performance is affected by its co-run applications, and the latter measures how it impacts the performance of its co-run applications. This paper shows that sensitivity and pressure are both application-and architecture dependent. Further, we propose a regression model that predicts an applications sensitivity and pressure across architectures with high accuracy. This regression model enables a data center scheduler to guarantee the QoS of a VM/application when it is scheduled to co-locate with another VMs/applications.

To feature space and back: Identifying top-weighted features in polynomial Support Vector Machine models

Polynomial Support Vector Machine models of degree d are linear functions in a feature space of monomials of at most degree d. However, the actual representation is stored in the form of support vectors and Lagrange multipliers that is unsuitable for human understanding. An efficient, heuristic method for searching the feature space of a polynomial Support Vector Machine model for those features with the largest absolute weights is presented. The time complexity of this method is Θdms^2 + sdp, where m is the number of variables, d the degree of the kernel, s the number of support vectors, and p the number of features the algorithm is allowed to search. In contrast, the brute force approach of constructing all weights and then selecting the largest weights has complexity Θsd{{m+d}\choose {d}}. The method is shown to be effective in identifying the top-weighted features on several simulated data sets, where the true weight vector is known. Additionally, the method is run on several high-dimensional, real world data sets where the features returned may be used to construct classifiers with classification performances similar to models built with all or subsets of variables returned by variable selection methods. This algorithm provides a new ability to understand, conceptualize, visualize, and communicate polynomial SVM models and has implications for feature construction, dimensionality reduction, and variable selection.

Selective switching mechanism in virtual machines via support vector machines and transfer learning

Virtualization is an essential technology in data centers allowing for a single machine to be used for multiple applications or users. With memory virtualization, two approaches, shadow paging (SP) and hardware-assisted paging (HAP), are taken by modern virtual machine memory managers. Neither memory mode is always preferred; previous studies have proposed to exploit the advantages of both modes by dynamically switching between these two paging modes based on the on-the-fly system behavior. However, the existing scheme makes the switching decision based on manual rules summarized for a specific architecture. This paper employs a machine learning approach that learns a decision model automatically and thus can adapt to different systems. Experimental results show that the performance of our switching mechanism can match or outperform either SP or HAP alone. Also, the results demonstrate that a machine learning-based decision model can match the performance of the hand-tuned model. Moreover, we further show that different hardware/software settings can affect on-the-fly system behavior and thus demand different decision models. Our scheme yields two effective decision models on two different machines. Additionally, transfer learning was used in order to efficiently train a model when faced with a new hardware configuration with only a limited number of training samples from the new machine.

Constructing Dynamic Policies for Paging Mode Selection

Virtualization technology is a key component for data center management which allows for multiple users and applications to share a single, physical machine. Modern virtual machine monitors utilize both software and hardware-assisted paging for memory virtualization, however neither paging mode is always preferable. Previous studies have shown that dynamic selection, which at runtime selects paging modes according to relevant performance metrics, can be effective in tailoring memory virtualization to program workload. However, these approaches require low-level manual analysis, or depend on prior knowledge of workload characteristics and phasing. We map the problem of dynamic paging mode selection to the contextual bandit, a model for sequential decision making in environments with limited feedback. Utilizing random profiling, which executes a workload while regularly selecting paging modes at random, we construct a paging mode selection policy that dynamically optimizes workload performance given page fault and translation lookaside buffer miss counts. Our approach yields an effective policy, DSP-OFFSET, for the dynamic paging mode selection problem. When trained and evaluated on subsets of the SPEC CPU2006 benchmark suite, DSP-OFFSET achieves speedups up to 44% compared to static paging mode selections, which is equivalent to the performance of the state-of-the-art ASP-SVM model. In addition, DSP-OFFSET requires at most a tenth of the profiling time of ASP-SVM (2.5 hours compared to over 24 hours) to achieve equivalent performance.

Lab exercises for a discrete structures course: exploring logic and relational algebra with Alloy

Students in computing disciplines need a strong basis in the fundamentals of discrete mathematics, but traditional offline approaches to teaching this material provide limited opportunities for the kind of interactive learning that computing students experience in their programming assignments. We have been using the Alloy language and analyzer to teach concepts in discrete structures (relational algebra, logic, and graphs) in an exploratory, programming-oriented way. Alloy, however, constitutes a new programming paradigm for introductory students, and careful mediation is needed to keep students on track. We use the familiar programming lab format, where students work on small-scope problems co-located with instructors, to provide guidance as students wrestle with the languages of relational algebra and predicate logic through Alloy. We describe selected lab exercises, and report on initial findings based on our experiences with students.