Search

Virtual Reality (VR) has emerged as a promising tool in many domains of therapy and rehabilitation, and has recently attracted the attention of researchers and clinicians working with elderly people with MCI, Alzheimer’s disease and related disorders. Here we present a study testing the feasibility of using highly realistic image-based rendered VR with patients with MCI and dementia. We designed an attentional task to train selective and sustained attention, and we tested a VR and a paper version of this task in a single-session within-subjects design. Results showed that participants with MCI and dementia reported to be highly satisfied and interested in the task, and they reported high feelings of security, low discomfort, anxiety and fatigue. In addition, participants reported a preference for the VR condition compared to the paper condition, even if the task was more difficult. Interestingly, apathetic participants showed a preference for the VR condition stronger than that of non-apathetic participants. These findings suggest that VR-based training can be considered as an interesting tool to improve adherence to cognitive training in elderly people with cognitive impairment. [ABSTRACT FROM AUTHOR]

Accurate Origin-Destination (OD) prediction is significant for effective traffic monitor, which can support operation decision in traffic planning and management field. The enclosed expressway network system like toll gates system in China can collect mounts of trip records which can be gathered for OD prediction. The paper develops a novel neural network, which is named Expressway OD Prediction Neural Network (EODPNN) for toll data-based prediction. The network consists of the following three modules: The Feature Extension Module, the Memory Module, and the Prediction Module. In the process, the attributes data which can reflect the city attribute such as GDP, population, and the number of vehicles are considered to embeded into the notwork to increase the accuracy of the model. For the applicability improvment of the model, we categorize the cities in multiple classes based on their economy and population scales in this paper, which can provide a higher accurate prediction of OD by EODPNN. The results shows that, comparing to the traditional model like ARIMA and SVM, or typical neural networks like Bidirectional Long Short-term Memory, the EODPNN delivers a better prediction performance. The method proposed in this paper has been fully verified and has a potential to transplant to the other OD data-based management systems for a more accurate and flexible prediction. [ABSTRACT FROM AUTHOR]

Open source software has been widely used in various industries due to its openness and flexibility, but it also brings potential security problems. Therefore, security analysis is required before using open source software. The current mainstream open source software vulnerability analysis technology is based on source code, and there are problems such as false positives, false negatives and restatements. In order to solve the problems, based on the further study of behavior feature extraction and vulnerability detection technology, a method of using dynamic behavior features to detect open source software vulnerabilities is proposed. Firstly, the relationship between open source software vulnerability and API call sequence is studied. Then, the behavioral risk vulnerability database of open source software is proposed as a support for vulnerability detection. In addition, the CNN-IndRNN classification model is constructed by improving the Independently Recurrent Neural Net-work (IndRNN) algorithm and applies to open source software security vulnerability detection. The experimental results verify the effectiveness of the proposed open source software security vulnerability detection method based on dynamic behavior features. [ABSTRACT FROM AUTHOR]

In this paper, we propose an Emotional Trigger System to impart an automatic emotion expression ability within the humanoid robot REN-XIN, in which the Emotional Trigger is an emotion classification model trained from our proposed Word Mover’s Distance(WMD) based algorithm. Due to the long time delay of the WMD-based Emotional Trigger System, we propose an enhanced Emotional Trigger System to enable a smooth interaction with the robot in which the Emotional Trigger is replaced by a conventional convolution neural network and a long short term memory network (CNN_LSTM)-based deep neural network. In our experiments, the CNN_LSTM based model only need 10 milliseconds or less to finish the classification without a decrease in accuracy, while the WMD-based model needed approximately 6-8 seconds to give a result. In this paper, the experiments are conducted based on the same sub-data sets of the Chinese emotional corpus(Ren_CECps) used in former WMD experiments: one comprises 50% data for training and 50% for testing(1v1 experiment), and the other comprises 80% data for training and 20% for testing(4v1 experiment). The experiments are conducted using WMD, CNN_LSTM, CNN and LSTM. The results show that CNN_LSTM obtains the best F1 score (0.35) in the 1v1 experiment and almost the same accuracy of F1 scores (0.366 vs 0.367) achieved by WMD in the 4v1 experiment. Finally, we present demonstration videos with the same scenario to show the performance of robot control driven by CNN_LSTM-based Emotional Trigger System and WMD-based Emotional Trigger System. To improve the comparison, total manual-control performance is also recorded. [ABSTRACT FROM AUTHOR]

To solve the problem of the deadline-constrained task scheduling in the cloud computing system, this paper proposes a deadline-constrained scheduling algorithm for cloud computing based on the driver of dynamic essential path (Deadline-DDEP). According to the changes of the dynamic essential path of each task node in the scheduling process, the dynamic sub-deadline strategy is proposed. The strategy assigns different sub-deadline values to every task node to meet the constraint relations among task nodes and the user’s defined deadline. The strategy fully considers the dynamic sub-deadline affected by the dynamic essential path of task node in the scheduling process. The paper proposed the quality assessment of optimization cost strategy to solve the problem of selecting server for each task node. Based on the sub-deadline urgency and the relative execution cost in the scheduling process, the strategy selects the server that not only meets the sub-deadline but also obtains much lower execution cost. In this way, the proposed algorithm will make the task graph complete within its deadline, and minimize its total execution cost. Finally, we demonstrate the proposed algorithm via the simulation experiments using Matlab tools. The experimental results show that, the proposed algorithm produces remarkable performance improvement rate on the total execution cost that ranges between 10.3% and 30.8% under meeting the deadline constraint. In view of the experimental results, the proposed algorithm provides better-quality scheduling solution that is suitable for scientific application task execution in the cloud computing environment than IC-PCP, DCCP and CD-PCP. [ABSTRACT FROM AUTHOR]

This paper studies the drive-response synchronization for quaternion-valued shunting inhibitory cellular neural networks (QVSICNNs) with mixed delays. First, QVSICNN is decomposed into an equivalent real-valued system in order to avoid the non-commutativity of the multiplicity. Then, the existence of almost periodic solutions is obtained based on the Banach fixed point theorem. An novel state-feedback controller is designed to ensure the global exponential almost periodic synchronization. At the end of the paper, an example is given to illustrate the effectiveness of the obtained results. [ABSTRACT FROM AUTHOR]

With the development of computer network bandwidth, packet classification algorithms which are able to deal with large-scale rule sets are in urgent need. Among the existing algorithms, researches on packet classification algorithms based on hierarchical trie have become an important packet classification research branch because of their widely practical use. Although hierarchical trie is beneficial to save large storage space, it has several shortcomings such as the existence of backtracking and empty nodes. This paper proposes a new packet classification algorithm, Hierarchical Trie Algorithm Based on Expectation-Maximization Clustering (HTEMC). Firstly, this paper uses the formalization method to deal with the packet classification problem by means of mapping the rules and data packets into a two-dimensional space. Secondly, this paper uses expectation-maximization algorithm to cluster the rules based on their aggregate characteristics, and thereby diversified clusters are formed. Thirdly, this paper proposes a hierarchical trie based on the results of expectation-maximization clustering. Finally, this paper respectively conducts simulation experiments and real-environment experiments to compare the performances of our algorithm with other typical algorithms, and analyzes the results of the experiments. The hierarchical trie structure in our algorithm not only adopts trie path compression to eliminate backtracking, but also solves the problem of low efficiency of trie updates, which greatly improves the performance of the algorithm. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of energy management of mixed criticality applications in a multi-core partitioned architecture. Instead of focusing on new scheduling algorithms to adjust frequency in order to save energy, we propose a partition to CPU allocation that takes into account not only the different frequencies at which the CPU can operate but the level of criticality of the partitions. The goal is to provide a set of pre-calculated allocations, called profiles, so at run time the system can switch to different modes depending on the battery level. These profiles achieve different levels of energy saving and performance applying different strategies. We also present a comparison in terms of energy saving of the most used bin-packing algorithms for partition allocation. As this is an heuristic, it is not possible to ensure that our results involve the minimum energy consumption. For this reason, we also provide a comparison with a exact method, such as constraint programming. [ABSTRACT FROM AUTHOR]

A recurrent criticism concerning the use of online social media data in political science research is the lack of demographic information about social media users. By employing a face-recognition algorithm to the profile pictures of Facebook users, the paper derives two fundamental demographic characteristics (age and gender) of a sample of Facebook users who interacted with the most relevant British parties in the two weeks before the Brexit referendum of 23 June 2016. The article achieves the goals of (i) testing the precision of the algorithm, (ii) testing its validity, (iii) inferring new evidence on digital mobilisation, and (iv) tracing the path for future developments and application of the algorithm. The findings show that the algorithm is reliable and that it can be fruitfully used in political and social sciences both to confirm the validity of survey data and to obtain information from populations that are generally unavailable within traditional surveys. [ABSTRACT FROM AUTHOR]

Specific memory might be stored in a subnetwork consisting of a small population of neurons. To select neurons involved in memory formation, neural competition might be essential. In this paper, we show that excitable neurons are competitive and organize into two assemblies in a recurrent network with spike timing-dependent synaptic plasticity (STDP) and axonal conduction delays. Neural competition is established by the cooperation of spontaneously induced neural oscillation, axonal conduction delays, and STDP. We also suggest that the competition mechanism in this paper is one of the basic functions required to organize memory-storing subnetworks into fine-scale cortical networks. [ABSTRACT FROM AUTHOR]

Textual representations play an important role in the field of natural language processing (NLP). The efficiency of NLP tasks, such as text comprehension and information extraction, can be significantly improved with proper textual representations. As neural networks are gradually applied to learn the representation of words and phrases, fairly efficient models of learning short text representations have been developed, such as the continuous bag of words (CBOW) and skip-gram models, and they have been extensively employed in a variety of NLP tasks. Because of the complex structure generated by the longer text lengths, such as sentences, algorithms appropriate for learning short textual representations are not applicable for learning long textual representations. One method of learning long textual representations is the Long Short-Term Memory (LSTM) network, which is suitable for processing sequences. However, the standard LSTM does not adequately address the primary sentence structure (subject, predicate and object), which is an important factor for producing appropriate sentence representations. To resolve this issue, this paper proposes the dependency-based LSTM model (D-LSTM). The D-LSTM divides a sentence representation into two parts: a basic component and a supporting component. The D-LSTM uses a pre-trained dependency parser to obtain the primary sentence information and generate supporting components, and it also uses a standard LSTM model to generate the basic sentence components. A weight factor that can adjust the ratio of the basic and supporting components in a sentence is introduced to generate the sentence representation. Compared with the representation learned by the standard LSTM, the sentence representation learned by the D-LSTM contains a greater amount of useful information. The experimental results show that the D-LSTM is superior to the standard LSTM for sentences involving compositional knowledge (SICK) data. [ABSTRACT FROM AUTHOR]

Approximate string matching with k-differences has a number of practical applications, ranging from pattern recognition to computational biology. This paper proposes an efficient memory-access algorithm for parallel approximate string matching with k-differences on Graphics Processing Units (GPUs). In the proposed algorithm, all threads in the same GPUs warp share data using warp-shuffle operation instead of accessing the shared memory. Moreover, we implement the proposed algorithm by exploiting the memory structure of GPUs to optimize its performance. Experiment results for real DNA packages revealed that the performance of the proposed algorithm and its implementation archived up to 122.64 and 1.53 times compared to that of sequential algorithm on CPU and previous parallel approximate string matching algorithm on GPUs, respectively. [ABSTRACT FROM AUTHOR]

Affective analysis of images in social networks has drawn much attention, and the texts surrounding images are proven to provide valuable semantic meanings about image content, which can hardly be represented by low-level visual features. In this paper, we propose a novel approach for visual affective classification (VAC) task. This approach combines visual representations along with novel text features through a fusion scheme based on Dempster-Shafer (D-S) Evidence Theory. Specifically, we not only investigate different types of visual features and fusion methods for VAC, but also propose textual features to effectively capture emotional semantics from the short text associated to images based on word similarity. Experiments are conducted on three public available databases: the International Affective Picture System (IAPS), the Artistic Photos and the MirFlickr Affect set. The results demonstrate that the proposed approach combining visual and textual features provides promising results for VAC task. [ABSTRACT FROM AUTHOR]

This paper proposes a pipelined non-deterministic finite automaton (NFA)-based string matching scheme using field programmable gate array (FPGA) implementation. The characteristics of the NFA such as shared common prefixes and no failure transitions are considered in the proposed scheme. In the implementation of the automaton-based string matching using an FPGA, each state transition is implemented with a look-up table (LUT) for the combinational logic circuit between registers. In addition, multiple state transitions between stages can be performed in a pipelined fashion. In this paper, it is proposed that multiple one-to-one state transitions, called merged state transitions, can be performed with an LUT. By cutting down the number of used LUTs for implementing state transitions, the hardware overhead of combinational logic circuits is greatly reduced in the proposed pipelined NFA-based string matching scheme. [ABSTRACT FROM AUTHOR]

Recently, cloud computing has drawn significant attention from both industry and academia, bringing unprecedented changes to computing and information technology. The infrastructure-as-a-Service (IaaS) model offers new abilities such as the elastic provisioning and relinquishing of computing resources in response to workload fluctuations. However, because the demand for resources dynamically changes over time, the provisioning of resources in a way that a given budget is efficiently utilized while maintaining a sufficing performance remains a key challenge. This paper addresses the problem of task scheduling and resource provisioning for a set of tasks running on IaaS clouds; it presents novel provisioning and scheduling algorithms capable of executing tasks within a given budget, while minimizing the slowdown due to the budget constraint. Our simulation study demonstrates a substantial reduction up to 70% in the overall task slowdown rate by the proposed algorithms. [ABSTRACT FROM AUTHOR]

Saliency detection attracted attention of many researchers and had become a very active area of research. Recently, many saliency detection models have been proposed and achieved excellent performance in various fields. However, most of these models only consider low-level features. This paper proposes a novel saliency detection model using both color and texture features and incorporating higher-level priors. The SLIC superpixel algorithm is applied to form an over-segmentation of the image. Color saliency map and texture saliency map are calculated based on the region contrast method and adaptive weight. Higher-level priors including location prior and color prior are incorporated into the model to achieve a better performance and full resolution saliency map is obtained by using the up-sampling method. Experimental results on three datasets demonstrate that the proposed saliency detection model outperforms the state-of-the-art models. [ABSTRACT FROM AUTHOR]

The wealth of sensory data coming from different modalities has opened numerous opportunities for data analysis. The data are of increasing volume, complexity and dimensionality, thus calling for new methodological innovations towards multimodal data processing. However, multimodal architectures must rely on models able to adapt to changes in the data distribution. Differences in the density functions can be due to changes in acquisition conditions (pose, illumination), sensors characteristics (number of channels, resolution) or different views (e.g. street level vs. aerial views of a same building). We call these different acquisition modes domains, and refer to the adaptation problem as domain adaptation. In this paper, instead of adapting the trained models themselves, we alternatively focus on finding mappings of the data sources into a common, semantically meaningful, representation domain. This field of manifold alignment extends traditional techniques in statistics such as canonical correlation analysis (CCA) to deal with nonlinear adaptation and possibly non-corresponding data pairs between the domains. We introduce a kernel method for manifold alignment (KEMA) that can match an arbitrary number of data sources without needing corresponding pairs, just few labeled examples in all domains. KEMA has interesting properties: 1) it generalizes other manifold alignment methods, 2) it can align manifolds of very different complexities, performing a discriminative alignment preserving each manifold inner structure, 3) it can define a domain-specific metric to cope with multimodal specificities, 4) it can align data spaces of different dimensionality, 5) it is robust to strong nonlinear feature deformations, and 6) it is closed-form invertible, which allows transfer across-domains and data synthesis. To authors’ knowledge this is the first method addressing all these important issues at once. We also present a reduced-rank version of KEMA for computational efficiency, and discuss the generalization performance of KEMA under Rademacher principles of stability. Aligning multimodal data with KEMA reports outstanding benefits when used as a data pre-conditioner step in the standard data analysis processing chain. KEMA exhibits very good performance over competing methods in synthetic controlled examples, visual object recognition and recognition of facial expressions tasks. KEMA is especially well-suited to deal with high-dimensional problems, such as images and videos, and under complicated distortions, twists and warpings of the data manifolds. A fully functional toolbox is available at . [ABSTRACT FROM AUTHOR]

In recent years, with the development of computer science, deep learning is held as competent enough to solve the problem of inference and learning in high dimensional space. Therefore, it has received unprecedented attention from both the academia and the business community. Compared with CPU/GPU, FPGA has attracted much attention for its high-energy efficiency, short development cycle and reconfigurability in the aspect of deep learning algorithm. However, because of the limited research on OpenCL optimization on FPGA of deep learning algorithms, OpenCL tools and models applied to CPU/GPU cannot be directly used on FPGA. This makes it difficult for software programmers to use FPGA when implementing deep learning algorithms for a rewarding performance. To solve this problem, this paper proposed an OpenCL computational model based on FPGA template architecture to optimize the time-consuming convolution layer in deep learning. The comparison between the program applying the computational model and the corresponding optimization program provided by Xilinx indicates that the former is 8-40 times higher than the latter in terms of performance. [ABSTRACT FROM AUTHOR]

This paper presents the Hybrid Scalable-Minimized-Butterfly-Fat-Tree (H-SMBFT) topology for on-chip communication. Main aspects of this work are the description of the architectural design and the characteristics as well as a comparative analysis against two established indirect topologies namely Butterfly-Fat-Tree (BFT) and Scalable-Minimized-Butterfly-Fat-Tree (SMBFT). Simulation results demonstrate that the proposed topology outperforms its predecessors in terms of performance, area and power dissipation. Specifically, it improves the link interconnectivity between routing levels, such that the number of required links isreduced. This results into reduced router complexity and shortened routing paths between any pair of communicating nodes in the network. Moreover, simulation results under synthetic as well as real-world embedded applications workloads reveal that H-SMBFT can reduce the average latency by up-to35.63% and 17.36% compared to BFT and SMBFT, respectively. In addition, the power dissipation of the network can be reduced by up-to33.82% and 19.45%, while energy consumption can be improved byup-to32.91% and 16.83% compared to BFT and SMBFT, respectively. [ABSTRACT FROM AUTHOR]

The path signature is a means of feature generation that can encode nonlinear interactions in data in addition to the usual linear terms. It provides interpretable features and its output is a fixed length vector irrespective of the number of input points or their sample times. In this paper we use the path signature to provide features for identifying people whose diagnosis subsequently converts to Alzheimer’s disease. In two separate classification tasks we distinguish converters from 1) healthy individuals, and 2) individuals with mild cognitive impairment. The data used are time-ordered measurements of the whole brain, ventricles and hippocampus from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). We find two nonlinear interactions which are predictive in both cases. The first interaction is change of hippocampal volume with time, and the second is a change of hippocampal volume relative to the volume of the whole brain. While hippocampal and brain volume changes are well known in Alzheimer’s disease, we demonstrate the power of the path signature in their identification and analysis without manual feature selection. Sequential data is becoming increasingly available as monitoring technology is applied, and the path signature method is shown to be a useful tool in the processing of this data. [ABSTRACT FROM AUTHOR]

Filter or screening methods are often used as a preprocessing step for reducing the number of variables used by a learning algorithm in obtaining a classification or regression model. While there are many such filter methods, there is a need for an objective evaluation of these methods. Such an evaluation is needed to compare them with each other and also to answer whether they are at all useful, or a learning algorithm could do a better job without them. For this purpose, many popular screening methods are partnered in this paper with three regression learners and five classification learners and evaluated on ten real datasets to obtain accuracy criteria such as R-square and area under the ROC curve (AUC). The obtained results are compared through curve plots and comparison tables in order to find out whether screening methods help improve the performance of learning algorithms and how they fare with each other. Our findings revealed that the screening methods were useful in improving the prediction of the best learner on two regression and two classification datasets out of the ten datasets evaluated. [ABSTRACT FROM AUTHOR]

Short-term metro passenger flow forecasting is an essential component of intelligent transportation systems (ITS) and can be applied to optimize the passenger flow organization of a station and offer data support for metro passenger flow early warning and system management. LSTM neural networks have recently achieved remarkable recent in the field of natural language processing (NLP) because they are well suited for learning from experience to predict time series. For this purpose, we propose an empirical mode decomposition (EMD)-based long short-term memory (LSTM) neural network model for predicting short-term metro inbound passenger flow. The EMD algorithm decomposes the original sequential passenger flow into several intrinsic mode functions (IMFs) and a residual. Selected IMFs that are strongly correlated with the original data can be obtained via feature selection. The selected IMFs and the original data are integrated into inputs for LSTM neural networks, and a single LSTM prediction model and an EMD-LSTM hybrid forecasting model are developed. Finally, historical real automatic fare collection (AFC) data from metro passengers are collected from Chengdu Metro to verify the validity of the proposed EMD-LSTM prediction model. The results indicate that the proposed EMD-LSTM hybrid forecasting model outperforms the LSTM, ARIMA and BPN models. [ABSTRACT FROM AUTHOR]

Background: High quality dietary intake data is required to support evidence of diet-disease relationships exposed in clinical research. Source data verification may be a useful quality assurance method in this setting. The present pilot study aimed to apply source data verification to evaluate the quality of the data coding process for dietary intake in a clinical trial and to explore potential barriers to data quality in this setting. Methods: Using a sample of 20 cases from a clinical trial, source data verification was conducted between three sets of data derived documents: transcripts of audio-recorded diet history interviews, matched paper-based diet history forms and outputs from nutrition analysis software. The number of cases and rates of discrepancies between documents were calculated. A total of five in-depth interviews with dietitians collecting and coding dietary data were thematically analysed. Results: Some 2024 discrepancies were identified. The highest discrepancy rate was 57.49%, and occurred between diet history interviews and nutrition analysis software outputs. Sources of the discrepancies included both quantities and frequencies of food intake. The highest discrepancy rate was for the food group “vegetable products and dishes”. In-depth interviews implicated recall bias of trial participants as a cause of discrepancies, but dietitians also acknowledged a possible subconscious influence of having to code reported foods into nutrition analysis software programs. Conclusion: The accuracy of dietary intake data appeared to depend on the level of detailed food data required. More support for participants on reporting consumption, and incorporating supportive tools to guide estimates of food quantities may facilitate a more consistent coding process and improve data quality. This pilot study offers a novel method and an overview of dietary intake data coding measurement errors. These findings may warrant further investigation in a larger sample. [ABSTRACT FROM AUTHOR]

SPEC CPU is one of the most common benchmark suites used in computer architecture research. CPU2017 has recently been released to replace CPU2006. In this paper we present a detailed evaluation of the memory hierarchy performance for both the CPU2006 and single-threaded CPU2017 benchmarks. The experiments were executed on an Intel Xeon Skylake-SP, which is the first Intel processor to implement a mostly non-inclusive last-level cache (LLC). We present a classification of the benchmarks according to their memory pressure and analyze the performance impact of different LLC sizes. We also test all the hardware prefetchers showing they improve performance in most of the benchmarks. After comprehensive experimentation, we can highlight the following conclusions: i) almost half of SPEC CPU benchmarks have very low miss ratios in the second and third level caches, even with small LLC sizes and without hardware prefetching, ii) overall, the SPEC CPU2017 benchmarks demand even less memory hierarchy resources than the SPEC CPU2006 ones, iii) hardware prefetching is very effective in reducing LLC misses for most benchmarks, even with the smallest LLC size, and iv) from the memory hierarchy standpoint the methodologies commonly used to select benchmarks or simulation points do not guarantee representative workloads. [ABSTRACT FROM AUTHOR]

Targeted environmental and ecosystem management remain crucial in control of dengue. However, providing detailed environmental information on a large scale to effectively target dengue control efforts remains a challenge. An important piece of such information is the extent of the presence of potential dengue vector breeding sites, which consist primarily of open containers such as ceramic jars, buckets, old tires, and flowerpots. In this paper we present the design and implementation of a pipeline to detect outdoor open containers which constitute potential dengue vector breeding sites from geotagged images and to create highly detailed container density maps at unprecedented scale. We implement the approach using Google Street View images which have the advantage of broad coverage and of often being two to three years old which allows correlation analyses of container counts against historical data from manual surveys. Containers comprising eight of the most common breeding sites are detected in the images using convolutional neural network transfer learning. Over a test set of images the object recognition algorithm has an accuracy of 0.91 in terms of F-score. Container density counts are generated and displayed on a decision support dashboard. Analyses of the approach are carried out over three provinces in Thailand. The container counts obtained agree well with container counts from available manual surveys. Multi-variate linear regression relating densities of the eight container types to larval survey data shows good prediction of larval index values with an R-squared of 0.674. To delineate conditions under which the container density counts are indicative of larval counts, a number of factors affecting correlation with larval survey data are analyzed. We conclude that creation of container density maps from geotagged images is a promising approach to providing detailed risk maps at large scale. [ABSTRACT FROM AUTHOR]

Computer-aided polyp detection in gastric gastroscopy has been the subject of research over the past few decades. However, despite significant advances, automatic polyp detection in real time is still an unsolved problem. In this paper, we report on a convolutional neural network (CNN) for polyp detection that is constructed based on Single Shot MultiBox Detector (SSD) architecture and which we call SSD for Gastric Polyps (SSD-GPNet). To take full advantages of feature maps’ information from the feature pyramid and to acquire higher accuracy, we re-use information that is abandoned by Max-Pooling layers. In other words, we reuse the lost data from the pooling layers and concatenate that data as extra feature maps to contribute to classification and detection. Meanwhile, in the feature pyramid, we concatenate feature maps of the lower layers and feature maps that are deconvolved from upper layers to make explicit relationships between layers and to effectively increase the number of channels. The results show that our enhanced SSD for gastric polyp detection can realize real-time polyp detection with 50 frames per second (FPS) and can improve the mean average precision (mAP) from 88.5% to 90.4%, with only a little loss in time-performance. And the further experiment shows that SSD-GPNet has excellent performance in improving polyp detection recalls over 10% (p = 0.00053), especially in small polyp detection. This can help endoscopic physicians more easily find missed polyps and decrease the gastric polyp miss rate. It may be applicable in daily clinical practice to reduce the burden on physicians. [ABSTRACT FROM AUTHOR]

Given a real-world graph, how can we measure relevance scores for ranking and link prediction? Random walk with restart (RWR) provides an excellent measure for this and has been applied to various applications such as friend recommendation, community detection, anomaly detection, etc. However, RWR suffers from two problems: 1) using the same restart probability for all the nodes limits the expressiveness of random walk, and 2) the restart probability needs to be manually chosen for each application without theoretical justification. We have two main contributions in this paper. First, we propose R W E R (RWER), a random walk based measure which improves the expressiveness of random walks by using a distinct restart probability for each node. The improved expressiveness leads to superior accuracy for ranking and link prediction. Second, we propose SR (pervised start for RWER), an algorithm for learning the restart probabilities of RWER from a given graph. SR eliminates the need to heuristically and manually select the restart parameter for RWER. Extensive experiments show that our proposed method provides the best performance for ranking and link prediction tasks. [ABSTRACT FROM AUTHOR]

Recognizing a person in a crowded environment is a challenging, yet critical, visual-search task for both humans and machine-vision algorithms. This paper explores the possibility of combining a residual neural network (ResNet), brain-computer interfaces (BCIs) and human participants to create “cyborgs” that improve decision making. Human participants and a ResNet undertook the same face-recognition experiment. BCIs were used to decode the decision confidence of humans from their EEG signals. Different types of cyborg groups were created, including either only humans (with or without the BCI) or groups of humans and the ResNet. Cyborg groups decisions were obtained weighing individual decisions by confidence estimates. Results show that groups of cyborgs are significantly more accurate (up to 35%) than the ResNet, the average participant, and equally-sized groups of humans not assisted by technology. These results suggest that melding humans, BCI, and machine-vision technology could significantly improve decision-making in realistic scenarios. [ABSTRACT FROM AUTHOR]

Effort-aware just-in-time (JIT) defect prediction is to rank source code changes based on the likelihood of detects as well as the effort to inspect such changes. Accurate defect prediction algorithms help to find more defects with limited effort. To improve the accuracy of defect prediction, in this paper, we propose a deep learning based approach for effort-aware just-in-time defect prediction. The key idea of the proposed approach is that neural network and deep learning could be exploited to select useful features for defect prediction because they have been proved excellent at selecting useful features for classification and regression. First, we preprocess ten numerical metrics of code changes, and then feed them to a neural network whose output indicates how likely the code change under test contains bugs. Second, we compute the benefit cost ratio for each code change by dividing the likelihood by its size. Finally, we rank code changes according to their benefit cost ratio. Evaluation results on a well-known data set suggest that the proposed approach outperforms the state-of-the-art approaches on each of the subject projects. It improves the average recall and popt by 15.6% and 8.1%, respectively. [ABSTRACT FROM AUTHOR]

Attention maps have been fused in the VggNet structure (EAC-Net) [] and have shown significant improvement compared to that of the VggNet structure. However, in [], E-Net was designed based on the facial action unit (AU) center and for facial AU detection only. Thus, for the use of attention maps in every image type, this paper proposed a new convolutional neural network (CNN) structure, P_VggNet, comprising the following parts: P_Net and VggNet with 16 layers (VggNet-16). The generation approach of P_Net was designed, and the P_VggNet structure was proposed. To prove the efficiency of P_VggNet, we designed two experiments, which indicated that P_VggNet could more efficiently extract image features than VggNet-16. [ABSTRACT FROM AUTHOR]

Background: Regular expression matching (REM) is widely employed as the major tool for deep packet inspection (DPI) applications. For automatic processing, the regular expression patterns need to be converted to a deterministic finite automata (DFA). However, with the ever-increasing scale and complexity of pattern sets, state explosion problem has brought a great challenge to the DFA based regular expression matching. Rule grouping is a direct method to solve the state explosion problem. The original rule set is divided into multiple disjoint groups, and each group is compiled to a separate DFA, thus to significantly restrain the severe state explosion problem when compiling all the rules to a single DFA. Objective: For practical implementation, the total number of DFA states should be as few as possible, thus the data structures of these DFAs can be deployed on fast on-chip memories for rapid access. In addition, to support fast pattern update in some applications, the time cost for grouping should be as small as possible. In this study, we aimed to propose an efficient grouping method, which generates as few states as possible with as little time overhead as possible. Methods: When compiling multiple patterns into a single DFA, the number of DFA states is usually greater than the total number of states when compiling each pattern to a separate DFA. This is mainly caused by the semantic overlaps among different rules. By quantifying the interaction values for each pair of rules, the rule grouping problem can be reduced to the maximum k-cut graph partitioning problem. Then, we propose a heuristic algorithm called the one-step greedy (OSG) algorithm to solve this NP-hard problem. What’s more, a subroutine named the heuristic initialization (HI) algorithm is devised to further optimize the grouping algorithms. Results: We employed three practical rule sets for the experimental evaluation. Results show that the OSG algorithm outperforms the state-of-the-art grouping solutions regarding both the total number of DFA states and time cost for grouping. The HI subroutine also demonstrates its significant optimization effect on the grouping algorithms. Conclusions: The DFA state explosion problem has became the most challenging issue in the regular expression matching applications. Rule grouping is a practical direction by dividing the original rule sets into multiple disjoint groups. In this paper, we investigate the current grouping solutions, and propose a compact and efficient grouping algorithm. Experiments conducted on practical rule sets demonstrate the superiority of our proposal. [ABSTRACT FROM AUTHOR]

This paper describes an approach for improving the accuracy of memory-based collaborative filtering, based on the technique for order of preference by similarity to ideal solution (TOPSIS) method. Recommender systems are used to filter the huge amount of data available online based on user-defined preferences. Collaborative filtering (CF) is a commonly used recommendation approach that generates recommendations based on correlations among user preferences. Although several enhancements have increased the accuracy of memory-based CF through the development of improved similarity measures for finding successful neighbors, there has been less investigation into prediction score methods, in which rating/preference scores are assigned to items that have not yet been selected by a user. A TOPSIS solution for evaluating multiple alternatives based on more than one criterion is proposed as an alternative to prediction score methods for evaluating and ranking items based on the results from similar users. The recommendation accuracy of the proposed TOPSIS technique is evaluated by applying it to various common CF baseline methods, which are then used to analyze the MovieLens 100K and 1M benchmark datasets. The results show that CF based on the TOPSIS method is more accurate than baseline CF methods across a number of common evaluation metrics. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new model of memristive multidirectional associative memory neural networks, which concludes the time-varying delays in leakage terms via sampled-data control. We use the input delay method to turn the sampling system into a continuous time-delaying system. Then we analyze the exponential stability and asymptotic stability of the equilibrium points for this model. By constructing a suitable Lyapunov function, using the Lyapunov stability theorem and some inequality techniques, some sufficient criteria for ensuring the stability of equilibrium points are obtained. Finally, numerical examples are given to demonstrate the effectiveness of our results. [ABSTRACT FROM AUTHOR]

As digital devices, such as smartphones, are becoming ever more absorbed in the daily lives of adolescents, a major assumption is that they start taking over basic functions of the human mind. A main focus of current debate and research is therefore on investigating adolescents’ use of digital technologies. However, the lack of an instrument measuring the degree to which adolescents offload cognitive and social functions to technology hinders debate and research. This paper tests the reliability and validity of the Extended Mind Questionnaire (XMQ) which measures the degree to which digital technology is used to offload cognitive and social functions. In a first study on young adults (n = 63), we constructed a 12-tem scale, which proved to be highly reliable. A large-scale study on teenagers (n = 947) demonstrated the high structural validity, reliability, and construct and criterion validity of the XMQ. In sum, these studies provide evidence that the XMQ is psychometrically sound and valid, and can be useful in future research on the consequences of digital technology in the daily lives of adolescents. [ABSTRACT FROM AUTHOR]

Sparse spectral clustering (SSC) has become one of the most popular clustering approaches in recent years. However, its high computational complexity prevents its application to large-scale datasets such as hyperspectral images (HSIs). In this paper, we propose two efficient approximate sparse spectral clustering methods for HSIs clustering in which clustering performance is improved by utilizing local information among the data. Firstly, we construct a smaller representative dataset on which sparse spectral clustering is performed. Then the labels of ground object are extending to whole dataset based on the local information according to two extending strategies. The first one is that the local interpolation is utilized to improve the extension of the clustering result. The other one is that the label extension is turned to a problem of subspace embedding, and is fulfilled by locally linear embedding (LLE). Several experiments on HSIs demonstrated that the proposed algorithms are effective for HSIs clustering. [ABSTRACT FROM AUTHOR]

Microprocessors in safety-critical system are extremely vulnerable to hacker attacks and circuit crosstalk, as they can modify binaries and lead programs to run along the wrong control flow paths. It is a significant challenge to design a run-time validation method with few hardware modification. In this paper, an efficient control flow validation method named DCM (Dual-Processor Control Flow Validation Method) is proposed basing on dual-processor architecture. Since a burst of memory-access-intensive instructions could block pipeline and cause lots of waiting clocks, the DCM assigns the idle pipeline cycles of the blocked processor to the other processor to validate control flow at run time. An extra lightweight monitor unit in each processor is needed and a special dual-processor communication protocol is also designed to schedule the redundant computing capacity between two processors to do validation tasks better. To further improve the efficiency, we also design a software-based self-validation algorithm to help reduce validation times. The combination of both hardware method and software method can speed up the validation procedure and protect the control flow paths with different emphasis. The cycle-accurate simulator GEM5 is used to simulate two ARMv7-A processors with out-of-order pipeline. Experiment shows the performance overhead of DCM is less than 22% on average across the SPEC 2006 benchmarks. [ABSTRACT FROM AUTHOR]

Recently, sparse representation, which relies on the underlying assumption that samples can be sparsely represented by their labeled neighbors, has been applied with great success to image classification problems. Through sparse representation-based classification (SRC), the label can be assigned with minimum residual between the sample and its synthetic version with class-specific coding, which means that the coding scheme is the most significant factor for classification accuracy. However, conventional SRC-based coding schemes ignore dependency among the samples, which leads to an undesired result that similar samples may be coded into different categories due to quantization sensitivity. To address this problem, in this paper, a novel approach based on self-supervised sparse representation is proposed for image classification. In the proposed approach, the manifold structure of samples is firstly exploited with low rank representation. Next, the low-rank representation matrix is used to characterize the similarity of samples in order to establish a self-supervised sparse coding model, which aims to preserve the local structure of codings for similar samples. Finally, a numerical algorithm utilizing the alternating direction method of multipliers (ADMM) is developed to obtain the approximate solution. Experiments on several publicly available datasets validate the effectiveness and efficiency of our proposed approach compared with existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The Bag-of-Visual-Words (BoVW) model is widely used for image classification, object recognition and image retrieval problems. In BoVW model, the local features are quantized and 2-D image space is represented in the form of order-less histogram of visual words. The image classification performance suffers due to the order-less representation of image. This paper presents a novel image representation that incorporates the spatial information to the inverted index of BoVW model. The spatial information is added by calculating the global relative spatial orientation of visual words in a rotation invariant manner. For this, we computed the geometric relationship between triplets of identical visual words by calculating an orthogonal vector relative to each point in the triplets of identical visual words. The histogram of visual words is calculated on the basis of the magnitude of these orthogonal vectors. This calculation provides the unique information regarding the relative position of visual words when they are collinear. The proposed image representation is evaluated by using four standard image benchmarks. The experimental results and quantitative comparisons demonstrate that the proposed image representation outperforms the existing state-of-the-art in terms of classification accuracy. [ABSTRACT FROM AUTHOR]

Facial reconstruction is a technique that aims to reproduce the individual facial characteristics based on interpretation of the skull, with the objective of recognition leading to identification. The aim of this paper was to evaluate the accuracy and recognition level of three-dimensional (3D) computerized forensic craniofacial reconstruction (CCFR) performed in a blind test on open-source software using computed tomography (CT) data from live subjects. Four CCFRs were produced by one of the researchers, who was provided with information concerning the age, sex, and ethnic group of each subject. The CCFRs were produced using Blender® with 3D models obtained from the CT data and templates from the MakeHuman® program. The evaluation of accuracy was carried out in CloudCompare, by geometric comparison of the CCFR to the subject 3D face model (obtained from the CT data). A recognition level was performed using the Picasa® recognition tool with a frontal standardized photography, images of the subject CT face model and the CCFR. Soft-tissue depth and nose, ears and mouth were based on published data, observing Brazilian facial parameters. The results were presented from all the points that form the CCFR model, with an average for each comparison between 63% and 74% with a distance -2.5 ≤ x ≤ 2.5 mm from the skin surface. The average distances were 1.66 to 0.33 mm and greater distances were observed around the eyes, cheeks, mental and zygomatic regions. Two of the four CCFRs were correctly matched by the Picasa® tool. Free software programs are capable of producing 3D CCFRs with plausible levels of accuracy and recognition and therefore indicate their value for use in forensic applications. [ABSTRACT FROM AUTHOR]

This paper proposes a shared synapse architecture for autoencoders (AEs), and implements an AE with the proposed architecture as a digital circuit on a field-programmable gate array (FPGA). In the proposed architecture, the values of the synapse weights are shared between the synapses of an input and a hidden layer, and between the synapses of a hidden and an output layer. This architecture utilizes less of the limited resources of an FPGA than an architecture which does not share the synapse weights, and reduces the amount of synapse modules used by half. For the proposed circuit to be implemented into various types of AEs, it utilizes three kinds of parameters; one to change the number of layers’ units, one to change the bit width of an internal value, and a learning rate. By altering a network configuration using these parameters, the proposed architecture can be used to construct a stacked AE. The proposed circuits are logically synthesized, and the number of their resources is determined. Our experimental results show that single and stacked AE circuits utilizing the proposed shared synapse architecture operate as regular AEs and as regular stacked AEs. The scalability of the proposed circuit and the relationship between the bit widths and the learning results are also determined. The clock cycles of the proposed circuits are formulated, and this formula is used to estimate the theoretical performance of the circuit when the circuit is used to construct arbitrary networks. [ABSTRACT FROM AUTHOR]

In this paper, we propose a face recognition algorithm based on a combination of vector quantization (VQ) and Markov stationary features (MSF). The VQ algorithm has been shown to be an effective method for generating features; it extracts a codevector histogram as a facial feature representation for face recognition. Still, the VQ histogram features are unable to convey spatial structural information, which to some extent limits their usefulness in discrimination. To alleviate this limitation of VQ histograms, we utilize Markov stationary features (MSF) to extend the VQ histogram-based features so as to add spatial structural information. We demonstrate the effectiveness of our proposed algorithm by achieving recognition results superior to those of several state-of-the-art methods on publicly available face databases. [ABSTRACT FROM AUTHOR]

Reactive navigation is a well-known paradigm for controlling an autonomous mobile robot, which suggests making all control decisions through some light processing of the current/recent sensor data. Among the many advantages of this paradigm are: 1) the possibility to apply it to robots with limited and low-priced hardware resources, and 2) the fact of being able to safely navigate a robot in completely unknown environments containing unpredictable moving obstacles. As a major disadvantage, nevertheless, the reactive paradigm may occasionally cause robots to get trapped in certain areas of the environment—typically, these conflicting areas have a large concave shape and/or are full of closely-spaced obstacles. In this last respect, an enormous effort has been devoted to overcome such a serious drawback during the last two decades. As a result of this effort, a substantial number of new approaches for reactive navigation have been put forward. Some of these approaches have clearly improved the way how a reactively-controlled robot can move among densely cluttered obstacles; some other approaches have essentially focused on increasing the variety of obstacle shapes and sizes that could be successfully circumnavigated; etc. In this paper, as a starting point, we choose the best existing reactive approach to move in densely cluttered environments, and we also choose the existing reactive approach with the greatest ability to circumvent large intricate-shaped obstacles. Then, we combine these two approaches in a way that makes the most of them. From the experimental point of view, we use both simulated and real scenarios of challenging complexity for testing purposes. In such scenarios, we demonstrate that the combined approach herein proposed clearly outperforms the two individual approaches on which it is built. [ABSTRACT FROM AUTHOR]

The traditional username/password or PIN based authentication scheme, which still remains the most popular form of authentication, has been proved insecure, unmemorable and vulnerable to guessing, dictionary attack, key-logger, shoulder-surfing and social engineering. Based on this, a large number of new alternative methods have recently been proposed. However, most of them rely on users being able to accurately recall complex and unmemorable information or using extra hardware (such as a USB Key), which makes authentication more difficult and confusing. In this paper, we propose a Digital Memories based user authentication scheme adopting homomorphic encryption and a public key encryption design which can protect users’ privacy effectively, prevent tracking and provide multi-level security in an Internet & IoT environment. Also, we prove the superior reliability and security of our scheme compared to other schemes and present a performance analysis and promising evaluation results. [ABSTRACT FROM AUTHOR]

Finite time synchronization, which means synchronization can be achieved in a settling time, is desirable in some practical applications. However, most of the published results on finite time synchronization don’t include delays or only include discrete delays. In view of the fact that distributed delays inevitably exist in neural networks, this paper aims to investigate the finite time synchronization of memristor-based Cohen-Grossberg neural networks (MCGNNs) with both discrete delay and distributed delay (mixed delays). By means of a simple feedback controller and novel finite time synchronization analysis methods, several new criteria are derived to ensure the finite time synchronization of MCGNNs with mixed delays. The obtained criteria are very concise and easy to verify. Numerical simulations are presented to demonstrate the effectiveness of our theoretical results. [ABSTRACT FROM AUTHOR]

High-utility sequential pattern mining (HUSPM) has become an important issue in the field of data mining. Several HUSPM algorithms have been designed to mine high-utility sequential patterns (HUPSPs). They have been applied in several real-life situations such as for consumer behavior analysis and event detection in sensor networks. Nonetheless, most studies on HUSPM have focused on mining HUPSPs in precise data. But in real-life, uncertainty is an important factor as data is collected using various types of sensors that are more or less accurate. Hence, data collected in a real-life database can be annotated with existing probabilities. This paper presents a novel pattern mining framework called high utility-probability sequential pattern mining (HUPSPM) for mining high utility-probability sequential patterns (HUPSPs) in uncertain sequence databases. A baseline algorithm with three optional pruning strategies is presented to mine HUPSPs. Moroever, to speed up the mining process, a projection mechanism is designed to create a database projection for each processed sequence, which is smaller than the original database. Thus, the number of unpromising candidates can be greatly reduced, as well as the execution time for mining HUPSPs. Substantial experiments both on real-life and synthetic datasets show that the designed algorithm performs well in terms of runtime, number of candidates, memory usage, and scalability for different minimum utility and minimum probability thresholds. [ABSTRACT FROM AUTHOR]

In recent years, safer and more reliable biometric methods have been developed. Apart from the need for enhanced security, the media and entertainment sectors have also been applying biometrics in the emerging market of user-adaptable objects/systems to make these systems more user-friendly. However, the complexity of some state-of-the-art biometric systems (e.g., iris recognition) or their high false rejection rate (e.g., fingerprint recognition) is neither compatible with the simple hardware architecture required by reduced-size devices nor the new trend of implementing smart objects within the dynamic market of the Internet of Things (IoT). It was recently shown that an individual can be recognized by extracting features from their electrocardiogram (ECG). However, most current ECG-based biometric algorithms are computationally demanding and/or rely on relatively large (several seconds) ECG samples, which are incompatible with the aforementioned application fields. Here, we present a computationally low-cost method (patent pending), including simple mathematical operations, for identifying a person using only three ECG morphology-based characteristics from a single heartbeat. The algorithm was trained/tested using ECG signals of different duration from the Physionet database on more than 60 different training/test datasets. The proposed method achieved maximal averaged accuracy of 97.450% in distinguishing each subject from a ten-subject set and false acceptance and rejection rates (FAR and FRR) of 5.710±1.900% and 3.440±1.980%, respectively, placing Beat-ID in a very competitive position in terms of the FRR/FAR among state-of-the-art methods. Furthermore, the proposed method can identify a person using an average of 1.020 heartbeats. It therefore has FRR/FAR behavior similar to obtaining a fingerprint, yet it is simpler and requires less expensive hardware. This method targets low-computational/energy-cost scenarios, such as tiny wearable devices (e.g., a smart object that automatically adapts its configuration to the user). A hardware proof-of-concept implementation is presented as an annex to this paper. [ABSTRACT FROM AUTHOR]

The availability of digital X-ray detectors, together with advances in reconstruction algorithms, creates an opportunity for bringing 3D capabilities to conventional radiology systems. The downside is that reconstruction algorithms for non-standard acquisition protocols are generally based on iterative approaches that involve a high computational burden. The development of new flexible X-ray systems could benefit from computer simulations, which may enable performance to be checked before expensive real systems are implemented. The development of simulation/reconstruction algorithms in this context poses three main difficulties. First, the algorithms deal with large data volumes and are computationally expensive, thus leading to the need for hardware and software optimizations. Second, these optimizations are limited by the high flexibility required to explore new scanning geometries, including fully configurable positioning of source and detector elements. And third, the evolution of the various hardware setups increases the effort required for maintaining and adapting the implementations to current and future programming models. Previous works lack support for completely flexible geometries and/or compatibility with multiple programming models and platforms. In this paper, we present FUX-Sim, a novel X-ray simulation/reconstruction framework that was designed to be flexible and fast. Optimized implementation for different families of GPUs (CUDA and OpenCL) and multi-core CPUs was achieved thanks to a modularized approach based on a layered architecture and parallel implementation of the algorithms for both architectures. A detailed performance evaluation demonstrates that for different system configurations and hardware platforms, FUX-Sim maximizes performance with the CUDA programming model (5 times faster than other state-of-the-art implementations). Furthermore, the CPU and OpenCL programming models allow FUX-Sim to be executed over a wide range of hardware platforms. [ABSTRACT FROM AUTHOR]

The traditional scale invariant feature transform (SIFT) method can extract distinctive features for image matching. However, it is extremely time-consuming in SIFT matching because of the use of the Euclidean distance measure. Recently, many binary SIFT (BSIFT) methods have been developed to improve matching efficiency; however, none of them is invariant to mirror reflection. To address these problems, in this paper, we present a horizontal or vertical mirror reflection invariant binary descriptor named MBR-SIFT, in addition to a novel image matching approach. First, 16 cells in the local region around the SIFT keypoint are reorganized, and then the 128-dimensional vector of the SIFT descriptor is transformed into a reconstructed vector according to eight directions. Finally, the MBR-SIFT descriptor is obtained after binarization and reverse coding. To improve the matching speed and accuracy, a fast matching algorithm that includes a coarse-to-fine two-step matching strategy in addition to two similarity measures for the MBR-SIFT descriptor are proposed. Experimental results on the UKBench dataset show that the proposed method not only solves the problem of mirror reflection, but also ensures desirable matching accuracy and speed. [ABSTRACT FROM AUTHOR]

In this paper, based on our previous multi-pattern uniform resource locator (URL) binary-matching algorithm called HEM, we propose an improved multi-pattern matching algorithm called MH that is based on hash tables and binary tables. The MH algorithm can be applied to the fields of network security, data analysis, load balancing, cloud robotic communications, and so on—all of which require string matching from a fixed starting position. Our approach effectively solves the performance problems of the classical multi-pattern matching algorithms. This paper explores ways to improve string matching performance under the HTTP protocol by using a hash method combined with a binary method that transforms the symbol-space matching problem into a digital-space numerical-size comparison and hashing problem. The MH approach has a fast matching speed, requires little memory, performs better than both the classical algorithms and HEM for matching fields in an HTTP stream, and it has great promise for use in real-world applications. [ABSTRACT FROM AUTHOR]

Two, partially interwoven, hot topics in the analysis and statistical modeling of neural data, are the development of efficient and informative representations of the time series derived from multiple neural recordings, and the extraction of information about the connectivity structure of the underlying neural network from the recorded neural activities. In the present paper we show that state-space clustering can provide an easy and effective option for reducing the dimensionality of multiple neural time series, that it can improve inference of synaptic couplings from neural activities, and that it can also allow the construction of a compact representation of the multi-dimensional dynamics, that easily lends itself to complexity measures. We apply a variant of the ‘mean-shift’ algorithm to perform state-space clustering, and validate it on an Hopfield network in the glassy phase, in which metastable states are largely uncorrelated from memories embedded in the synaptic matrix. In this context, we show that the neural states identified as clusters’ centroids offer a parsimonious parametrization of the synaptic matrix, which allows a significant improvement in inferring the synaptic couplings from the neural activities. Moving to the more realistic case of a multi-modular spiking network, with spike-frequency adaptation inducing history-dependent effects, we propose a procedure inspired by Boltzmann learning, but extending its domain of application, to learn inter-module synaptic couplings so that the spiking network reproduces a prescribed pattern of spatial correlations; we then illustrate, in the spiking network, how clustering is effective in extracting relevant features of the network’s state-space landscape. Finally, we show that the knowledge of the cluster structure allows casting the multi-dimensional neural dynamics in the form of a symbolic dynamics of transitions between clusters; as an illustration of the potential of such reduction, we define and analyze a measure of complexity of the neural time series. [ABSTRACT FROM AUTHOR]