Memory Clusters Research Articles

Incorporating LSTM Auto-Encoders in Optimizations to Solve Parking Officer Patrolling Problem

The smart parking system is one of the most important problems in smart cities. Recently, an increasing number of sensors installed in parking spaces have provided big spatio-temporal data that be used to analyze parking situations in the city and help parking officers monitor parking violations. The traveling officer problem was customized to formulate a path-finding problem that aims to maximize the probability of catching overstayed cars before they leave. One of the challenges is to extract effective features from the big spatio-temporal data and provide a data-driven solution to replace conventional solutions such as a simple rule-based system or single optimization methods. In this article, we propose a seamless end-to-end learning and optimization framework that combines the long short-term memory auto-encoder neural network, clustering, and path-finding methods to solve the traveling officer problem. Our extensive comparison experiments on a large-scale real-world dataset have shown that our proposed solution outperforms any other single-step or optimization methods.

A Distributed Shared Memory Middleware for Speculative Parallel Discrete Event Simulation

The large diffusion of multi-core machines has pushed the research in the field of Parallel Discrete Event Simulation (PDES) toward new programming paradigms, based on the exploitation of shared memory. On the opposite side, the advent of Cloud computing—and the possibility to group together many (low-cost) virtual machines to form a distributed memory cluster capable of hosting simulation applications—has raised the need to bridge shared memory programming and seamless distributed execution. In this article, we present the design of a distributed middleware that transparently allows a PDES application coded for shared memory systems to run on clusters of (Cloud) resources. Our middleware is based on a synchronization protocol called Event and Cross State Synchronization. It allows cross-simulation-object access by event handlers, thus representing a powerful tool for the development of various types of PDES applications. We also provide data for an experimental assessment of our middleware architecture, which has been integrated into the open source ROOT-Sim speculative PDES platform.

КОЛЛЕКТИВНАЯ ПАМЯТЬ О СОВЕТСКОМ ПРОШЛОМ НА YOUTUBE: СПОСОБЫ АУДИОВИЗУАЛЬНОГО КОНСТРУИРОВАНИЯ РЯДОВЫМИ ВИДЕОБЛОГЕРАМИ

The YouTube platform hosts a multitude of video clips that contain various media representations of the collective memory of Soviet history. In this array of multimedia products, videos created by ordinary bloggers take their own place. Videos with themes about the recent Past are formed on the basis of digitized fragments of documentary and fiction films, photographs, posters, postcards, drawings, etc. Verbal comments of viewers-users to videos expand and change their semantic content. The purpose of the article is to reveal in the videos of ordinary bloggers various techniques for audiovisual constructing the collective memory of the mass repression of the 1930s and the Gulag. The criteria for selecting sources for the study are: the time they were posted on the YouTube platform (2010s), video bloggers claim to be documentary, the number of comments. The research results demonstrate that bloggers shape clusters of collective memory of the repressions and the Gulag using audiovisual signs and symbols that are easily recognizable by viewers. The same photographs and film fragments can “work” in video differently depending on how they relate to the verbal and audio components of multimodal text. The techniques used by bloggers for audiovisual documentation of the Soviet Past combine pre-digital and digital technologies. The selection of certain audiovisual components as “raw data” for video, their montage, giving historical figures certain roles, organizing time and space in a digital narrative, including different social and political contexts in it, all allow bloggers to form various representations of the collective memory of the Soviet past, depending on their ideological positions.

Moderate levels of physical fitness maintain telomere length in non-senescent T CD8+ cells of aged men

OBJECTIVES:Immunosenescence is an age-associated change characterized by a decreased immune response. Although physical activity has been described as fundamental for maintaining the quality of life, few studies have evaluated the effects of different levels of exercise on telomere length in aged populations. The present study aimed to analyze the effects of different levels of physical activity, classified by the Maximal oxygen consumption (VO2 max) values, on the telomere length of memory Cluster of differentiation (CD) CD4+(CD45ROneg and CD45RO+), effector CD8+CD28neg, and CD8+CD28+ T cells in aged individuals.METHODS:Fifty-three healthy elderly men (aged 65-85 years) were included in this study. Their fitness level was classified according to the American College of Sports Medicine (ACSM) for VO2 max (mL/kg/min). Blood samples were obtained from all participants to analyze the percentage of CD3, CD4, CD8, CD28+, naïve, and subpopulations of memory T cells by using flow cytometry. Furthermore, using the Flow-FISH methodology, the CD4+CD45RO+, CD4+CD45ROneg, CD8+CD28+, and CD8+CD28negT cell telomere lengths were measured.RESULTS:There was a greater proportion of effector memory T CD4+ cells and longer telomeres in CD8+CD28+ T cells in the moderate physical fitness group than in the other groups. There was a higher proportion of terminally differentiated memory effector T cells in the low physical fitness group.CONCLUSION:A moderate physical activity may positively influence the telomere shortening of CD28+CD8+T cells. However, additional studies are necessary to evaluate the importance of this finding with regard to immune function responses in older men.

Identification and Validation of Predictive Biomarkers to CD19- and BCMA-Specific CAR T-Cell Responses in CAR T-Cell Precursors

Identification and Validation of Predictive Biomarkers to CD19- and BCMA-Specific CAR T-Cell Responses in CAR T-Cell Precursors

Performance analysis of synchronous and asynchronous distributed genetic algorithms on multiprocessors

Abstract Because of their effectiveness and flexibility in finding useful solutions, Genetic Algorithms (GAs) are very popular search techniques for solving complex optimization problems in scientific and industrial fields. Parallel GAs (PGAs), and especially distributed ones have been usually presented as the way to overcome the time-consuming shortcoming of sequential GAs. In the case of applying PGAs, we can expect better performance, the reason being the exchange of knowledge during the parallel search process. The resulting distributed search is different compared to what sequential panmictic GAs do, then deserving additional studies. This article presents a performance study of three different PGAs. Moreover, we investigate the effect of synchronizing communications over modern shared-memory multiprocessors. We consider the master-slave model along with synchronous and asynchronous distributed GAs (dGAs), presenting their different designs and expected similarities when running in a number of cores ranging from one to 32 cores. The master-slave model showed a competitive numerical effort versus the other dGAs and demonstrated to be able to scale-up well over multiprocessors. We describe how the speed-up and parallel performance of the dGAs is changing as the number of cores enlarges. Results of the island model show that synchronous and asynchronous dGAs have different numerical performances on a multiprocessor, the asynchronous algorithm having a faster execution, thus more attractive for time demanding applications. Our results and statistical analyses help in developing a novel body of knowledge on PGAs running in shared memory multiprocessors (versus overwhelming literature oriented to distributed memory clusters), something useful for researchers, beginners, and final users of these techniques.

Projected memory clustering

We present a new algorithm PMC (Projected Memory Clustering) for projected clustering of high dimensional data. It effectively discovers clusters described by affine subspaces parallel to the main axes of coordinate system. The number of clusters and dimensions of subspaces are selected automatically. Experiments performed on various types of data show that PMC detects clustering structures better than related projected clustering methods. Moreover, it is fast, which makes it suitable for practical use.

Big Data Performance Analysis on a Hadoop Distributed File System Based on Geometric Data Perturbation Technique

Abstract In this paper, we proposed a Big Data Security Analysis based on a modified perturbation technique in a Hadoop Distributed File System which is commonly used in various real time applications. This work have utilized an improved K-means clustering algorithm, which selects the initial clustering centres based on the density parameters. The poor assurance of initial centres and in order to improve the precision and packing effect of the K-means clustering computation is to be need a new method. In order to group the data of same cluster based on their similarities clustering algorithm was used. In the various clustering methods, this paper is utilizing the geometric data perturbation technique which is clustered, data are perturbed whose values are challenging to be recognized. The performance was evaluated on a sample Health Care database and the metrics under studies are memory usage, precision, recall, accuracy, F-measure, clustering time and execution time, Perturbation time, De-perturbation time. In this paper, we focused only on Perturbation Time, De-Perturbation Time and the Accuracy. The experimental result shows that the proposed approach reduces the complexity and shown to less computation time and better accuracy than that of the existing techniques.

Change point dynamics for financial data: an indexed Markov chain approach

This paper uses an Indexed Markov Chain to model high frequency price returns of quoted rms. Introducing an Index process permits consideration of endogenous market volatility, and two important stylized facts of financial time series can be taken into account: long memory and volatility clustering. This paper rst proposes a method to optimally determine the state space of the Index process, which is based on a change-point approach for Markov chains. Furthermore, we provide an explicit formula for the probability distribution function of the rst change of state of the Index process. Results are illustrated with an application to intra-day firm prices.

Explicit memory based ABC with a clustering strategy for updating and retrieval of memory in dynamic environments

The Artificial Bee Colony (ABC) algorithm is considered as one of the swarm intelligence optimization algorithms. It has been extensively used for the applications of static type. Many practical and real-world applications, nevertheless, are of dynamic type. Thus, it is needed to employ some optimization algorithms that could solve this group of the problems that are of dynamic type. Dynamic optimization problems in which change(s) may occur through the time are tougher to face than static optimization problems. In this paper, an approach based on the ABC algorithm enriched with explicit memory and population clustering scheme, for solving dynamic optimization problems is proposed. The proposed algorithm uses the explicit memory to store the aging best solutions and employs clustering for preserving diversity in the population. Using the aging best solutions and keeping the diversity in population of the candidate solutions in the environment help speed-up the convergence of the algorithm. The proposed approach has been tested on Moving Peaks Benchmark. The Moving Peaks Benchmark is a suitable function for testing optimization algorithms and it is considered as one of the best representative of dynamic environments. The experimental study on the Moving Peaks Benchmark shows that the proposed approach is superior to several other well-known and state-of-the-art algorithms in dynamic environments.

The effects of bilingual language proficiency on recall accuracy and semantic clustering in free recall output: evidence for shared semantic associations across languages

ABSTRACTTwo experiments investigated how well bilinguals utilise long-standing semantic associations to encode and retrieve semantic clusters in verbal episodic memory. In Experiment 1, Spanish-English bilinguals (N = 128) studied and recalled word and picture sets. Word recall was equivalent in L1 and L2, picture recall was better in L1 than in L2, and the picture superiority effect was stronger in L1 than in L2. Semantic clustering in word and picture recall was equivalent in L1 and L2. In Experiment 2, Spanish-English bilinguals (N = 128) and English-speaking monolinguals (N = 128) studied and recalled word sequences that contained semantically related pairs. Data were analyzed using a multinomial processing tree approach, the pair-clustering model. Cluster formation was more likely for semantically organised than for randomly ordered word sequences. Probabilities of cluster formation, cluster retrieval, and retrieval of unclustered items did not differ across languages or language groups. Language proficiency has little if any impact on the utilisation of long-standing semantic associations, which are language-general.

Reversibility of peripheral blood leukocyte phenotypic and functional changes after exposure to and withdrawal from tofacitinib, a Janus kinase inhibitor, in healthy volunteers

This study evaluated the short-term effects of tofacitinib treatment on peripheral blood leukocyte phenotype and function, and the reversibility of any such effects following treatment withdrawal in healthy volunteers. Cytomegalovirus (CMV)-seropositive subjects received oral tofacitinib 10 mg twice daily for 4 weeks and were followed for 4 weeks after drug withdrawal. There were slight increases in total lymphocyte and total T-cell counts during tofacitinib treatment, and B-cell counts increased by up to 26%. There were no significant changes in granulocyte or monocyte counts, or granulocyte function. Naïve and central memory T-cell counts increased during treatment, while all subsets of activated T cells were decreased by up to 69%. T-cell subsets other than effector memory cluster of differentiation (CD)4+, activated naïve CD4+ and effector CD8+ T-cell counts and B-cell counts, normalized 4 weeks after withdrawal. Following ex vivo activation, measures of CMV-specific T-cell responses, and antigen non-specific T-cell-mediated cytotoxicity and interferon (IFN)-γ production, decreased slightly. These T-cell functional changes were most pronounced at Day 15, partially normalized while still on tofacitinib and returned to baseline after drug withdrawal. Total natural killer (NK)-cell counts decreased by 33%, returning towards baseline after drug withdrawal. NK-cell function decreased during tofacitinib treatment, but without a consistent time course across measured parameters. However, markers of NK-cell-mediated cytotoxicity, antibody-dependent cellular cytotoxicity and IFN-γ production were decreased up to 42% 1 month after drug withdrawal. CMV DNA was not detectable in whole blood, and there were no cases of herpes zoster reactivation. No new safety concerns arose. In conclusion, the effect of short-term tofacitinib treatment on leukocyte composition and function in healthy CMV+ volunteers is modest and largely reversible 4 weeks after withdrawal.

Lossy compression approach to subspace clustering

We present a novel subspace clustering algorithm SuMC (Subspace Memory Clustering) based on information theory, MDLP (Minimal Description Length Principle) and lossy compression. SuMC simultaneously solves two fundamental problems of subspace clustering: determination of the number of clusters and their optimal dimensions.Although SuMC requires only two parameters: data compression ratio r and a number of bits that are used to code a single scalar, the optimal value of compression ratio can be estimated by the Bayesian information criterion (BIC).We verified that in typical tasks of clustering, image segmentation and data compression, we obtain either better or comparable results to the leading methods of subspace clustering.

A hybrid parallel Delaunay image-to-mesh conversion algorithm scalable on distributed-memory clusters

In this paper, we present a scalable three dimensional parallel Delaunay image-to-mesh conversion algorithm. A nested master–worker model is used to simultaneously explore process- and thread-level parallelization. The mesh generation includes two stages: coarse and fine meshing. First, a coarse mesh is constructed in parallel by the threads of the master process. Then the coarse mesh is partitioned. Finally, the fine mesh refinement procedure is executed until all the elements in the mesh satisfy the quality and fidelity criteria. The communication and computation are separated during the fine mesh refinement procedure. The master thread of each process that initializes the MPI environment is in charge of the inter-node MPI communication for data (submesh) movement while the worker threads of each process are responsible for the local mesh refinement within the node. We conducted a set of experiments to test the performance of the algorithm on distributed memory clusters and observed that the granularity of coarse level data decomposition, which affects the coarse level concurrency, has a significant influence on the performance of the algorithm. With the proper value of granularity, the algorithm is scalable to 45 distributed memory compute nodes (900 cores).

Theory of Mind and Executive Functions are Dissociated in Multiple Sclerosis.

To investigate the relationship of executive functions (EF) and theory of mind (ToM) in multiple sclerosis (MS), clarifing whether ToM impairment in MS is related to EF dysfunction or whether they represent dissociable processes which can be independently impaired in MS. We enrolled 60 consecutive MS patients and 60 healthy controls (HC) matched on age, gender, and education level. All participants underwent ToM testing using the Eyes Test and the Videos Test and neuropsychological testing tapping different EF subdomains. A correlation analysis and a hierarchical cluster analysis were used to determine the similarity between explained variance between EF measures and ToM tests. MS patients had lower scores on both tasks of ToM compared to HC, i.e., Eyes Test (21.1 ± 5.0 vs. 29.5 ± 3.8, p < .001; Cohen's d: 1.9) and Videos Test (19.6 ± 2.4 vs. 22.9 ± 1.8, p < .001; Cohen's d: 1.6). They also performed significantly worse on different measures of EF. ToM performance was not significantly correlated with EF tests. The hierarchical cluster analysis showed that ToM measures in MS were clustered separately from the EF measures, revealing three executive clusters (Attention/working memory cluster; Inhibitory control/shifting ability cluster; Verbal Initiative/Abstract reasoning cluster) and one ToM cluster. This study suggests a dissociation of EF and ToM in MS, meaning that the MS-related neurobehavioral symptoms may be associated with a significant impairment in ToM independent of the level of EF performance. Ultimately, this discrimination of ToM deficits in MS may help to identify the appropriate cognitive and behavioral interventions.

$$\hbox {TM}^{2}$$ TM 2 C: a software transactional memory for many-cores

Transactional memory is an appealing paradigm for concurrent systems. Many software implementations of the paradigm were proposed in the past two decades for both shared memory multi-core systems and clusters of distributed machines. Chip manufacturers have however started producing many-core architectures, with low network-on-chip communication latencies and limited support for cache coherence, rendering existing transactional-memory implementations inapplicable. This paper presents $$\hbox {TM}^{2}\hbox {C}$$ , the first software transactional memory protocol for many-core systems, hence featuring transactions that are both distributed and leverage shared memory. $$\hbox {TM}^{2}\hbox {C}$$ exploits fast messages over network-on-chip to make accesses to shared data coherent. In particular, it allows visible read accesses to detect conflicts eagerly and incorporates the first distributed contention manager that guarantees the commit of all transactions. We evaluate $$\hbox {TM}^{2}\hbox {C}$$ on Intel, AMD and Tilera architectures, ranging from common multi-cores to experimental many-cores. We build upon new message-passing protocols, based on both software and hardware, which are interesting in their own right. Our results on various benchmarks, including realistic banking and MapReduce applications, show that $$\hbox {TM}^{2}\hbox {C}$$ scales well regardless of the underlying platform.

NEURODEGENERATIVE PATTERNS OF COGNITIVE CLUSTERS OF EARLY ONSET AD SUBJECTS: EVIDENCE FOR DISEASE HETEROGENEITY

Symptoms of Alzheimer's disease (AD) occur before the age of 65 (EO) in approximately 5% of patients. EOAD can profoundly affect non-memory domains early on and often time in isolation. We analyzed patterns of impairment in ADNI amnestic EOAD individuals using data-driven statistical analyses. The cognitive data of 146 non-familial EO ADNI subjects was Z-normalized to the performance of 395 cognitively normal (CN) individuals. Domain-averaged Z-scores were produced and corrected for age, gender and education. Wald cluster analysis on the residuals was performed. We compared the MRI, FDG-PET and amyloid-PET data of each identified EO cluster to 200 age-matched ADNI CN using linear regression in SPM8 while controlling for age, gender, and education. Additionally, intracranial volume and scan type were controlled for MRI. FDR/FWE correction for multiple comparisons was applied. The EOAD sample had mean age of 64.2±5.4 years, education of 16.1±2.8 years and MMSE 25.6±2.9. There were 49% male and 67% APOE ε4 carriers. Scree plot using the pseudo T-squared suggested a 4-cluster solution (see Table 1). Cluster 1 (n=64) showed a memory-predominant impairment and neurodegenerative changes (FDG>MRI) in the medial and lateral temporal, lateral parietal and posterior cingulate (see Figures 1 and 2). Cluster 2 (n=31) showed impaired memory and visuospatial function, and neurodegeneration (MRI>FDG) of the medial temporal and temporoparietal cortices and frontal lobes (MRI only). Cluster 3 (n=37), who were impaired in memory, language, and executive function and cluster 4 (n=14), who had impairment across all 4 domains, manifested neurodegeneration (MRI>FDG) throughout the brain. All clusters showed global amyloid burden (Figure 3).

NEURODEGENERATIVE PATTERNS OF COGNITIVE CLUSTERS OF EARLY ONSET AD SUBJECTS: EVIDENCE FOR DISEASE HETEROGENEITY

Symptoms of Alzheimer's disease (AD) occur before the age of 65 (EO) in approximately 5% of patients. EOAD can profoundly affect non-memory domains early on and often time in isolation. We analyzed patterns of impairment in ADNI amnestic EOAD individuals using data-driven statistical analyses. The cognitive data of 146 non-familial EO ADNI subjects was Z-normalized to the performance of 395 cognitively normal (CN) individuals. Domain-averaged Z-scores were produced and corrected for age, gender and education. Wald cluster analysis on the residuals was performed. We compared the MRI, FDG-PET and amyloid-PET data of each identified EO cluster to 200 age-matched ADNI CN using linear regression in SPM8 while controlling for age, gender, and education. Additionally, intracranial volume and scan type were controlled for MRI. FDR/FWE correction for multiple comparisons was applied. The EOAD sample had mean age of 64.2±5.4 years, education of 16.1±2.8 years and MMSE 25.6±2.9. There were 49% male and 67% APOE ε4 carriers. Scree plot using the pseudo T-squared suggested a 4-cluster solution (see Table 1). Cluster 1 (N=64) showed a memory-predominant impairment and neurodegenerative changes (FDG>MRI) in the medial and lateral temporal, lateral parietal and posterior cingulate (see Figures 1 and 2). Cluster 2 (N=31) showed impaired memory and visuospatial function, and neurodegeneration (MRI>FDG) of the medial temporal and temporoparietal cortices and frontal lobes (MRI only). Cluster 3 (N=37), who were impaired in memory, language, and executive function and cluster 4 (N=14), who had impairment across all 4 domains, manifested neurodegeneration (MRI>FDG) throughout the brain. All clusters showed global amyloid burden (Figure 3).

Neurocognitive Correlates of Attention-Deficit Hyperactivity Disorder Symptoms in Children Born at Extremely Low Gestational Age.

Compared with children born near term, those born extremely preterm (EP) are at much higher risk for attention-deficit hyperactivity disorder (ADHD). Little information is available about differences in neuropsychological outcomes among EP children with and without ADHD. Our analyses aimed to evaluate the neuropsychological correlates of ADHD symptoms in extremely low gestational age newborns (ELGANs). We obtained Child Symptom Inventory-4 reports from parents (n = 871) and teachers (n = 634) of 10-year-old children born before the 28th week of gestation. Participants completed standardized assessments of neurocognitive and academic functioning. In the total sample, children who screened positive for ADHD symptoms were at increased risk for neurocognitive limitations. These associations were weaker when the sample was limited to those with intelligence quotient (IQ) ≥70 or ≥85. Even those with IQ ≥85 who screened positive for ADHD symptoms were more likely than their peers to have deficits on the DAS-II Working Memory Cluster and the NEPSY-II Auditory Response subtest. The risks for impaired academic performance (Z ≤ -1) on components of the WIAT-III were 2-to-3 times higher in this group than among ELGANs not classified as having ADHD symptoms. Among children born EP, those with ADHD symptoms are more likely to have global neurocognitive impairment. When IQ is within normal limits, ADHD symptoms are associated with deficits in executive functioning skills. These findings highlight a group at risk for executive functioning deficits and related academic difficulties, even in the absence of intellectual disability.

Design and Implementation of a Communication-Optimal Classifier for Distributed Kernel Support Vector Machines

We consider the problem of how to design and implement communication-efficient versions of parallel kernel support vector machines, a widely used classifier in statistical machine learning, for distributed memory clusters and supercomputers. The main computational bottleneck is the training phase, in which a statistical model is built from an input data set. Prior to our study, the parallel isoefficiency of a state-of-the-art implementation scaled as $W=\Omega (P^3)$ , where $W$ is the problem size and $P$ the number of processors; this scaling is worse than even a one-dimensional block row dense matrix vector multiplication, which has $W=\Omega (P^2)$ . This study considers a series of algorithmic refinements, leading ultimately to a Communication-Avoiding SVM method that improves the isoefficiency to nearly $W=\Omega (P)$ . We evaluate these methods on 96 to 1,536 processors, and show average speedups of $3-16\times$ ( $7\times$ on average) over Dis-SMO, and a 95 percent weak-scaling efficiency on six real-world datasets, with only modest losses in overall classification accuracy. The source code can be downloaded at [1] .