Results 1 
8 of
8
Modern continuous optimization algorithms for tuning real and integer algorithm parameters
 LNCS 6234. Proceedings of the International Conference on Swarm Intelligence (ANTS 2010
, 2010
"... Abstract. To obtain peak performance from optimization algorithms, it is required to set appropriately their parameters. Frequently, algorithm parameters can take values from the set of real numbers, or from a large integer set. To tune this kind of parameters, it is interesting to apply stateofth ..."
Abstract

Cited by 10 (3 self)
 Add to MetaCart
Abstract. To obtain peak performance from optimization algorithms, it is required to set appropriately their parameters. Frequently, algorithm parameters can take values from the set of real numbers, or from a large integer set. To tune this kind of parameters, it is interesting to apply stateoftheart continuous optimization algorithms instead of using a tedious, and errorprone, handson approach. In this paper, we study the performance of several continuous optimization algorithms for the algorithm parameter tuning task. As case studies, we use a number of optimization algorithms from the swarm intelligence literature. 1
Understanding the Scalability of Bayesian Network Inference using Clique Tree Growth Curves
"... Bayesian networks (BNs) are used to represent and ef ciently compute with multivariate probability distributions in a wide range of disciplines. One of the main approaches to perform computation in BNs is clique tree clustering and propagation. In this approach, BN computation consists of propagati ..."
Abstract

Cited by 2 (2 self)
 Add to MetaCart
(Show Context)
Bayesian networks (BNs) are used to represent and ef ciently compute with multivariate probability distributions in a wide range of disciplines. One of the main approaches to perform computation in BNs is clique tree clustering and propagation. In this approach, BN computation consists of propagation in a clique tree compiled from a Bayesian network. There is a lack of understanding of how clique tree computation time, and BN computation time in more general, depends on variations in BN size and structure. On the one hand, complexity results tell us that many interesting BN queries are NPhard or worse to answer, and it is not hard to nd application BNs where the clique tree approach in practice cannot be used. On the other hand, it is wellknown that treestructured BNs can be used to answer probabilistic queries in polynomial time. In this article, we develop an approach to characterizing clique tree growth as a function of parameters that can be computed in polynomial time from BNs, speci cally: (i) the ratio of the number of a BN's nonroot nodes to the number of root nodes, or (ii) the expected number of moral edges in their moral graphs. Our approach is based on combining analytical and experimental results. Analytically, we partition the set of cliques in a clique tree into different sets, and introduce a growth curve for each set. For the special case of bipartite BNs, we consequently have two growth curves, a mixed clique growth curve and a root clique growth curve. In experiments, we systematically increase the degree of the root nodes in bipartite Bayesian networks, and nd that root clique growth is wellapproximated by Gompertz growth curves. It is believed that this research improves the understanding of the scaling behavior of clique tree clustering, provides a foundation for benchmarking and developing improved BN inference and machine learning algorithms, and presents an aid for analytical tradeoff studies of clique tree clustering using growth curves.
Portfolios in Stochastic Local Search: Efficiently Computing Most Probable Explanations in Bayesian Networks
"... In this article we investigate the use of portfolios (or collections) of heuristics when solving computationally hard problems using stochastic local search. We consider uncertainty reasoning, specifically the computation of most probable explanations in Bayesian networks (BNs). Our contribution is ..."
Abstract
 Add to MetaCart
(Show Context)
In this article we investigate the use of portfolios (or collections) of heuristics when solving computationally hard problems using stochastic local search. We consider uncertainty reasoning, specifically the computation of most probable explanations in Bayesian networks (BNs). Our contribution is twofold. First, we introduce a portfoliobased stochastic local search approach that utilizes an initialization portfolio and a search portfolio. Second, and leveraging our portfolio approach, we investigate determinism in BNs. We introduce a novel additive measure of gain (or gradient), which is tailored to partly deterministic Bayesian networks, and show how it generalizes the MAXSAT measure of gain from stochastic local search for the satisfiability (SAT) problem. This measure of gain provides, along with an explanation’s gain in probability and their respective noisy variants, different heuristics that locally improve an explanation. Our approach is implemented in a stochastic local search system, Stochastic Greedy Search. Stochastic Greedy Search is here compared to the stateoftheart inference system H����, which performs BN inference by compilation to and propagation in clique trees. We report on experiments using partly deterministic Bayesian networks from applications as well as synthetically generated networks. On synthetic
An Investigation of StateSpace Model Fidelity for SSME Data
"... Abstract—In previous studies, a variety of unsupervised anomaly detection techniques for anomaly detection were applied to SSME (Space Shuttle Main Engine) data. The observed results indicated that the identification of certain anomalies were specific to the algorithmic method under consideration. T ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract—In previous studies, a variety of unsupervised anomaly detection techniques for anomaly detection were applied to SSME (Space Shuttle Main Engine) data. The observed results indicated that the identification of certain anomalies were specific to the algorithmic method under consideration. This is the reason why one of the followon goals of these previous investigations was to build an architecture to support the best capabilities of all algorithms. We appeal to that goal here by investigating a cascade, serial architecture for the best performing and most suitable candidates from previous studies. As a precursor to a formal ROC (Receiver Operating Characteristic) curve analysis for validation of resulting anomaly detection algorithms, our primary focus here is to investigate the model fidelity as measured by variants of the AIC (Akaike Information Criterion) for statespace based models. We show that placing constraints on a statespace model during or after the training of the model introduces a modest level of suboptimality. Furthermore, we compare the fidelity of all candidate models including those embodying the cascade, serial architecture. We make recommendations on the most suitable candidates for application to subsequent anomaly detection studies as measured by AICbased criteria. I.
Efficient Probabilistic Diagnostics for Electrical Power Systems
"... We consider in this work the probabilistic approach to modelbased dignosis when applied to electrical power systems (EPSs). Our probabilistic approach is formally wellfounded, as it based on Bayesian networks and arithmetic circuits. We investigate the diagnostic task known as fault isolation, and ..."
Abstract
 Add to MetaCart
We consider in this work the probabilistic approach to modelbased dignosis when applied to electrical power systems (EPSs). Our probabilistic approach is formally wellfounded, as it based on Bayesian networks and arithmetic circuits. We investigate the diagnostic task known as fault isolation, and pay special attention to meeting two of the main challenges — model development and realtime reasoning — often associated with realworld application of modelbased diagnosis technologies. To address the challenge of model development, we develop a systematic approach to representing electrical power systems as Bayesian networks, supported by an easytouse specification language. To address the realtime reasoning challenge, we compile Bayesian networks into arithmetic circuits. Arithmetic circuit evaluation supports realtime diagnosis by being predictable and fast. In essence, we introduce a highlevel EPS specification language from which Bayesian networks that can diagnose multiple simultaneous failures are autogenerated, and we illustrate the feasibility of using arithmetic circuits, compiled from Bayesian networks, for realtime diagnosis on realworld EPSs of interest to NASA. The experimental system is a realworld EPS, namely the Advanced Diagnostic and Prognostic Testbed (ADAPT) located at the NASA Ames Research Center. In experiments with the ADAPT Bayesian network, which currently contains 503 discrete nodes and 579 edges, we find high diagnostic accuracy in scenarios where one to three faults, both in components and sensors, were inserted. The time taken to compute the most probable explanation using arithmetic circuits has a small mean of 0.2625 milliseconds and standard deviation of 0.2028 milliseconds. In experiments with data from ADAPT we also show that arithmetic circuit evaluation substantially outperforms joint tree propagation and variable elimination, two alternative algorithms for diagnosis using Bayesian network inference.
Analyzing Evolutionary Optimization in Noisy Environments
"... Many optimization tasks must be handled in noisy environments, where the exact evaluation of a solution cannot be obtained, only a noisy one. For optimization of noisy tasks, evolutionary algorithms (EAs), a type of stochastic metaheuristic search algorithm, have been widely and successfully applied ..."
Abstract
 Add to MetaCart
(Show Context)
Many optimization tasks must be handled in noisy environments, where the exact evaluation of a solution cannot be obtained, only a noisy one. For optimization of noisy tasks, evolutionary algorithms (EAs), a type of stochastic metaheuristic search algorithm, have been widely and successfully applied. Previous work mainly focuses on the empirical study and design of EAs for optimization under noisy conditions, while the theoretical understandings are largely insufficient. In this study, we firstly investigate how noisy fitness can affect the running time of EAs. Two kinds of noisehelpful problems are identified, on which the EAs will run faster with the presence of noise, and thus the noise should not be handled. Secondly, on a representative noiseharmful problem in which the noise has a strong negative effect, we examine two commonly employed mechanisms dealing with noise in EAs: the reevaluation and the threshold selection strategies. The analysis discloses that using these two strategies simultaneously is effective for the onebit noise, but ineffective for the asymmetric onebit noise. The smooth threshold selection is then proposed, which can be proven as an effective strategy to further improve the noise tolerance ability in the problem. We then complement the theoretical analysis by experiments on both synthetic problems as well as two combinatorial problems, the minimum spanning tree and the maximum matching. The experimental results agree with the theoretical findings, and also show that the proposed smooth threshold selection can deal with the noise better.
Towards Software Health Management with Bayesian Networks Position Paper
"... More and more systems (e.g., aircraft, machinery, cars) rely heavily on software, which performs safetycritical operations. Assuring software safety though traditional V&V has become a tremendous, if not impossible task, given the growing size and complexity of the software. We propose that i ..."
Abstract
 Add to MetaCart
More and more systems (e.g., aircraft, machinery, cars) rely heavily on software, which performs safetycritical operations. Assuring software safety though traditional V&V has become a tremendous, if not impossible task, given the growing size and complexity of the software. We propose that iSWHM (Integrated SoftWare Health Management) can increase safety and reliability of highassurance software systems. iSWHM uses advanced techniques from the area of system health management in order to continuously monitor the behavior of the software during operation, quickly detect anomalies and perform automatic and reliable rootcause analysis, while not replacing traditional V&V. Information provided by the iSWHM system can be used for automatic mitigation mechanisms (e.g., recovery, dynamic reconfiguration) or presented to a human operator. iSWHM’s prognostic capabilities will further improve reliability and availability as it provides information about soontooccur failures or looming performance bottlenecks. In this paper, we will discuss challenges and future potential and describe how Bayesian networks (BN) could be used for iSWHM modeling.
Reactive Bayesian Network Computation using Feedback Control: An Empirical Study
"... This paper investigates the challenge of integrating intelligent systems into varying computational platforms and application mixes while providing reactive (or soft realtime) response. We integrate Bayesian network computation with feedback control, thereby achieving our reactive objective. As a c ..."
Abstract
 Add to MetaCart
(Show Context)
This paper investigates the challenge of integrating intelligent systems into varying computational platforms and application mixes while providing reactive (or soft realtime) response. We integrate Bayesian network computation with feedback control, thereby achieving our reactive objective. As a case study we investigate fault diagnosis using Bayesian networks. While we consider the likelihood weighting and junction tree propagation Bayesian network inference algorithms in some detail, we hypothesize that the techniques developed can be broadly applied to achieve reactive intelligent systems. In this paper’s empirical study we demonstrate reactive fault diagnosis for an electrical power system. 1