MQTT is one of the promising protocols for various data exchange in IoT environments. Typically, those environments have a characteristic called “edge-heavy”, which means that things at the network edge generate a massive volume of data with high locality. For handling such edge-heavy data, an architecture of placing multiple MQTT brokers at the network edges and making them cooperate with each other is quite effective. It can provide higher throughput and lower latency, as well as reducing consumption of cloud resources. However, under this kind of architecture, heterogeneity could be a vital issue. Namely, an appropriate product of MQTT broker could vary according to the different environment of each network edge, even though different products are hard to cooperate due to the MQTT specification providing no interoperability between brokers. In this paper, we propose Interworking Layer of Distributed MQTT brokers (ILDM), which enables arbitrary kinds of MQTT brokers to cooperate with each other. ILDM, designed as a generic mechanism independent of any specific cooperation algorithm, provides APIs to facilitate development of a variety of algorithms. By using the APIs, we also present two basic cooperation algorithms. To evaluate the usefulness of ILDM, we introduce a benchmark system which can be used for both a single broker and multiple brokers. Experimental results show that the throughput of five brokers running together by ILDM is improved 4.3 times at maximum than that of a single broker.

An annealing processor based on the Ising model is a remarkable candidate for combinatorial optimization problems and it is superior to general von Neumann computers. CMOS-based implementations of the annealing processor are efficient and feasible based on current semiconductor technology. However, critical problems with annealing processors remain. There are few simulated spins and inflexibility in terms of implementable graph topology due to hardware constraints. A prior approach to overcoming these problems is to emulate a complicated graph on a simple and high-density spin array with so-called minor embedding, a spin duplication method based on graph theory. When a complicated graph is embedded on such hardware, numerous spins are consumed to represent high-degree spins by combining multiple low-degree spins. In addition to the number of spins, the quality of solutions decreases as a result of dummy strong connections between the duplicated spins. Thus, the approach cannot handle large-scale practical problems. This paper proposes a flexible and scalable hardware architecture with time-division multiplexing for massive spins and high-degree topologies. A target graph is separated and mapped onto multiple virtual planes, and each plane is subject to interleaved simulation with time-division processing. Therefore, the behavior of high-degree spins is efficiently emulated over time, so that no dummy strong connections are required, and the solution quality is accordingly improved. We implemented a prototype hardware design for FPGAs, and we evaluated the proposed method in a software-based annealing processor simulator. The results indicate that the method increased the spins that can be deployed. In addition, our time-division multiplexing architecture improved the solution quality and convergence time with reasonable resource consumption.

We propose a novel compositing pipeline and a dynamic load balancing technique for volume rendering which utilizes a two-layered group structure to achieve effective and scalable load balancing. The technique enables each process to render data from non-contiguous regions of the volume with minimal impact on the total render time. We demonstrate the effectiveness of the proposed technique by performing a set of experiments on a modern GPU cluster. The experiments show that using the technique results in up to a 35.7% lower worst-case memory usage as compared to a dynamic k-d tree load balancing technique, whilst simultaneously achieving similar or higher render performance. The proposed technique was also able to lower the amount of transferred data during the load balancing stage by up to 72.2%. The technique has the potential to be used in many scenarios where other dynamic load balancing techniques have proved to be inadequate, such as during large-scale visualization.

In wireless multi-hop networks such as ad hoc networks and sensor networks, backoff-based opportunistic routing protocols, which make a forwarding decision based on backoff time, have been proposed. In the protocols, each potential forwarder calculates the backoff time based on the product of a weight and global scaling factor. The weight prioritizes potential forwarders and is calculated based on hop counts to the destination of a sender and receiver. The global scaling factor is a predetermined value to map the weight to the actual backoff time. However, there are three common issues derived from the global scaling factor. First, it is necessary to share the predetermined global scaling factor with a centralized manner among all terminals properly for the backoff time calculation. Second, it is almost impossible to change the global scaling factor during the networks are being used. Third, it is difficult to set the global scaling factor to an appropriate value since the value differs among each local surrounding of forwarders. To address the aforementioned issues, this paper proposes a novel decentralized local scaling factor control without relying on a predetermined global scaling factor. The proposed method consists of the following three mechanisms: (1) sender-centric local scaling factor setting mechanism in a decentralized manner instead of the global scaling factor, (2) adaptive scaling factor control mechanism which adapts the local scaling factor to each local surrounding of forwarders, and (3) mitigation mechanism for excessive local scaling factor increases for the local scaling factor convergence. Finally, this paper evaluates the backoff-based opportunistic routing protocol with and without the proposed method using computer simulations.

In this paper, we propose an acceleration method for the Held-Karp algorithm that solves the symmetric traveling salesman problem by dynamic programming. The proposed method achieves acceleration with two techniques. First, we locate data-independent subproblems so that the subproblems can be solved in parallel. Second, we reduce the number of subproblems by a meet in the middle (MITM) technique, which computes the optimal path from both clockwise and counterclockwise directions. We show theoretical analysis on the impact of MITM in terms of the time and space complexities. In experiments, we compared the proposed method with a previous method running on a single-core CPU. Experimental results show that the proposed method on an 8-core CPU was 9.5-10.5 times faster than the previous method on a single-core CPU. Moreover, the proposed method on a graphics processing unit (GPU) was 30-40 times faster than that on an 8-core CPU. As a side effect, the proposed method reduced the memory usage by 48%.

Deep neural network (NN) has been widely accepted for enabling various AI applications, however, the limitation of computational and memory resources is a major problem on mobile devices. Quantized NN with a reduced bit precision is an effective solution, which relaxes the resource requirements, but the accuracy degradation due to its numerical approximation is another problem. We propose a novel quantized NN model employing the “dithering” technique to improve the accuracy with the minimal additional hardware requirement at the view point of the hardware-algorithm co-designing. Dithering distributes the quantization error occurring at each pixel (neuron) spatially so that the total information loss of the plane would be minimized. The experiment we conducted using the software-based accuracy evaluation and FPGA-based hardware resource estimation proved the effectiveness and efficiency of the concept of an NN model with dithering.

The latency and the energy consumption of DRAM are serious concerns because (1) the latency has not improved much for decades and (2) recent machines have huge capacity of main memory. Device-level studies reduce them by shortening the wait time of DRAM internal operations so that they finish fast and consume less energy. Applying these techniques aggressively to achieve approximate memory is a promising direction to further reduce the overhead, given that many data-center applications today are to some extent robust to bit-flips. To advance research on approximate memory, it is required to evaluate its effect to applications so that both researchers and potential users of approximate memory can investigate how it affects realistic applications. However, hardware simulators are too slow to run workloads repeatedly with different parameters. To this end, we propose a lightweight method to evaluate effect of approximate memory. The idea is to count the number of DRAM internal operations that occur to approximate data of applications and calculate the probability of bit-flips based on it, instead of using heavy-weight simulators. The evaluation shows that our system is 3 orders of magnitude faster than cycle accurate simulators, and we also give case studies of evaluating effect of approximate memory to some realistic applications.

In statistical analysis and data mining, change-point detection that identifies the change-points which are times when the probability distribution of time series changes has been used for various purposes, such as anomaly detections on network traffic and transaction data. However, computation cost of a conventional AR (Auto-Regression) model based approach is too high and infeasible for online. In this paper, an AR model based online change-point detection algorithm, called ChangeFinder, is implemented on an FPGA (Field Programmable Gate Array) based NIC (Network Interface Card). The proposed system computes the change-point score from time series data received from 10GbE (10Gbit Ethernet). More specifically, it computes the change-point score at the 10GbE NIC in advance of host applications. It can find change-points on single or multiple streams using a context memory. This paper aims to reduce the host workload and improve change-point detection performance by offloading ChangeFinder algorithm from host to the NIC. As evaluations, change-point detection in the FPGA NIC is compared with a baseline software implementation and those enhanced by two network optimization techniques using DPDK and Netfilter in terms of throughput. The result demonstrates 16.8x improvement in change-point detection throughput compared to the baseline software implementation. It is corresponding to the 10GbE line rate. Performance and area overheads when supporting multiple streams are also evaluated.

Non-volatile memory (NVM) is a promising technology for low-energy and high-capacity main memory of computers. The characteristics of NVM devices, however, tend to be fundamentally different from those of DRAM (i.e., the memory device currently used for main memory), because of differences in principles of memory cells. Typically, the write latency of an NVM device such as PCM and ReRAM is much higher than its read latency. The asymmetry in read/write latencies likely affects the performance of applications significantly. For analyzing behavior of applications running on NVM-based main memory, most researchers use software-based emulation tools due to the limited number of commercial NVM products. However, these existing emulation tools are too slow to emulate a large-scale, realistic workload or too simplistic to investigate the details of application behavior on NVM with asymmetric read/write latencies. This paper therefore proposes a new NVM emulation mechanism that is not only light-weight but also aware of a read/write latency gap in NVM-based main memory. We implemented the prototype of the proposed mechanism for the Intel CPU processors of the Haswell architecture. We also evaluated its accuracy and performed case studies for practical benchmarks. The results showed that our prototype accurately emulated write-latencies of NVM-based main memory: it emulated the NVM write latencies in a range from 200 ns to 1000 ns with negligible errors from 0.2% to 1.1%. We confirmed that the use of our emulator enabled us to successfully estimate performance of practical workloads for NVM-based main memory, while an existing light-weight emulation model misestimated.

This paper proposes and examines the three in-memory shuffling methods designed to address problems in MapReduce shuffling caused by skewed data. Coupled Shuffle Architecture (CSA) employs a single pairwise all-to-all exchange to shuffle both blocks, units of shuffle transfer, and meta-blocks, which contain the metadata of corresponding blocks. Decoupled Shuffle Architecture (DSA) separates the shuffling of meta-blocks and blocks, and applies different all-to-all exchange algorithms to each shuffling process, attempting to mitigate the impact of stragglers in strongly skewed distributions. Decoupled Shuffle Architecture with Skew-Aware Meta-Shuffle (DSA w/ SMS) autonomously determines the proper placement of blocks based on the memory consumption of each worker process. This approach targets extremely skewed situations where some worker processes could exceed their node memory limitation. This study evaluates implementations of the three shuffling methods in our prototype in-memory MapReduce engine, which employs high performance interconnects such as InfiniBand and Intel Omni-Path. Our results suggest that DSA w/ SMS is the only viable solution for extremely skewed data distributions. We also present a detailed investigation of the performance of CSA and DSA in various skew situations.

The bulk execution of a sequential algorithm is to execute it for many different inputs in turn or at the same time. It is known that the bulk execution of an oblivious sequential algorithm can be implemented to run efficiently on a GPU. The bulk execution supports fine grained bitwise parallelism, allowing it to achieve high acceleration over a straightforward sequential computation. The main contribution of this work is to present a Bitwise Parallel Bulk Computation (BPBC) to accelerate the Smith-Waterman Algorithm (SWA) using the affine gap penalty. Thus, our idea is to convert this computation into a circuit simulation using the BPBC technique to compute multiple instances simultaneously. The proposed BPBC technique for the SWA has been implemented on the GPU and CPU. Experimental results show that the proposed BPBC for the SWA accelerates the computation by over 646 times as compared to a single CPU implementation and by 6.9 times as compared to a multi-core CPU implementation with 160 threads.

A set of spanning trees of a graphs G are called completely independent spanning trees (CISTs for short) if for every pair of vertices x, y∈V(G), the paths joining x and y in any two trees have neither vertex nor edge in common, except x and y. Constructing CISTs has applications on interconnection networks such as fault-tolerant routing and secure message transmission. In this paper, we investigate the problem of constructing two CISTs in the balanced hypercube BH_n, which is a hypercube-variant network and is superior to hypercube due to having a smaller diameter. As a result, the diameter of CISTs we constructed equals to 9 for BH₂ and 6n-2 for BH_n when n≥3.

Over the past recent decades, numerous programming languages have expanded to embrace multi-paradigms such as the fusion of object-oriented and functional programming. For example, Java, one of the most famous object-oriented programming languages, introduced a number of functional idioms in 2014. This evolution enables developers to achieve various benefits from both paradigms. However, we do not know how Java developers use functional idioms actually. Additionally, the extent to which, while there are several criticisms against the idioms, the developers actually accept and/or use the idioms currently remains unclear. In this paper, we investigate the actual use status of three functional idioms (Lambda Expression, Stream, and Optional) in Java projects by mining 100 projects containing approximately 130,000 revisions. From the mining results, we determined that Lambda Expression is utilized in 16% of all the examined projects, whereas Stream and Optional are only utilized in 2% to 3% of those projects. It appears that most Java developers avoid using functional idioms just because of keeping compatibility Java versions, while a number of developers accept these idioms for reasons of readability and runtime performance improvements. Besides, when they adopt the idioms, Lambda Expression frequently consists of a single statement, and Stream is used to operate the elements of a collection. On the other hand, some developers implement Optional using deprecated methods. We can say that good usage of the idioms should be widely known among developers.

Modern code review is a well-known practice to assess the quality of software where developers discuss the quality in a web-based review tool. However, this lightweight approach may risk an inefficient review participation, especially when comments becomes either excessive (i.e., too many) or underwhelming (i.e., too few). In this study, we investigate the phenomena of reviewer commenting. Through a large-scale empirical analysis of over 1.1 million reviews from five OSS systems, we conduct an exploratory study to investigate the frequency, size, and evolution of reviewer commenting. Moreover, we also conduct a modeling study to understand the most important features that potentially drive reviewer comments. Our results find that (i) the number of comments and the number of words in the comments tend to vary among reviews and across studied systems; (ii) reviewers change their behaviours in commenting over time; and (iii) human experience and patch property aspects impact the number of comments and the number of words in the comments.

Free/libre open source software (FLOSS) are being rapidly employed in several companies and organizations, because it can be modified and used for free. Hence, the use of FLOSS could contribute to its originally intended benefits and to the competence of its users. In this study, we analyzed the effect of using FLOSS on related competences. We investigated the change in the competences through an empirical study before and after the use of FLOSS among project participants. Consequently, it was confirmed that the competences of the participants improved after utilizing FLOSS.

Recently, many studies have applied gamification to software engineering education and software development to enhance work results. Gamification is defined as “the use of game design elements in non-game contexts.” When applying gamification, we make various game rules, such as a time limit. However, it is not clear whether the rule affects working time or not. For example, if we apply a time limit to impatient developers, the working time may become shorter, but the rule may negatively affect because of pressure for time. In this study, we analyze with subjective experiments whether the rules affects work results such as working time. Our experimental results suggest that for the coding tasks, working time was shortened when we applied a rule that made developers aware of working time by showing elapsed time.

Indexing is one of the fields where the non-volatile memory (NVM) has the advantages of byte-addressable characteristics and fast read/write speed. The existing index structures for NVM have been developed based on the fact that the size of cache line and the atomicity guarantee unit of NVM are different and they tried to overcome the weakness of consistency from the difference. To overcome the weakness, an expensive flush operation is required which results in a lower performance than a basic B+tree index. Recent studies have shown that the I/O units of the NVM can be matched with the atomicity guarantee units under limited circumstances. In this paper, we propose a Cache line sized Atomic Write B+tree (CAWBT), which is a minimal B+tree structure that shows higher performance than a basic b+ tree and designed for NVM. CAWBT has almost same performance compared to basic B+tree without consistency guarantee and shows remarkable performance improvement compared to other B+tree indexes for NVM.

File systems based on persistent memory deploy Copy-on-Write (COW) or logging to guarantee data consistency. However, COW has a write amplification problem and logging has a double write problem. Both COW and logging increase write traffic on persistent memory. In this work, we present adaptive differential logging and zero-copy logging for persistent memory. Adaptive differential logging applies COW or logging selectively to each block. If the updated size of a block is smaller than or equal to half of the block size, we apply logging to the block. If the updated size of a block is larger than half of the block size, we apply COW to the block. Zero-copy logging treats an user buffer on persistent memory as a redo log. Zero-copy logging does not incur any additional data copy. We implement adaptive differential logging and zero-copy logging on both NOVA and PMFS file systems. Our measurement on real workloads shows that adaptive differential logging and zero-copy logging get 150.6% and 149.2% performance improvement over COW, respectively.

Detecting phishing websites is imperative. Among several detection schemes, the promising ones are the visual similarity-based approaches. In those, targeted legitimate website's visual features referred to as signatures are stored in SDB (Signature Database) by the system administrator. They can only detect phishing websites whose signatures are highly similar to SDB's one. Thus, the system administrator has to register multiple signatures to detect various phishing websites and that cost is very high. This incurs the vulnerability of zero-day phishing attack. In order to address this issue, an auto signature update mechanism is needed. The naive way of auto updating SDB is expanding the scope of detection by adding detected phishing website's signature to SDB. However, the previous approaches are not suitable for auto updating since their similarity can be highly different among targeted legitimate website and subspecies of phishing website targeting that legitimate website. Furthermore, the previous signatures can be easily manipulated by attackers. In order to overcome the problems mentioned above, in this paper, we propose a hue signature auto update system for visual similarity-based phishing detection with tolerance to zero-day attack. The phishing websites targeting certain legitimate website tend to use the targeted website's theme color to deceive users. In other words, the users can easily distinguish phishing website if it has highly different hue information from targeted legitimate one (e.g. red colored Facebook is suspicious). Thus, the hue signature has a common feature among the targeted legitimate website and subspecies of phishing websites, and it is difficult for attackers to change it. Based on this notion, we argue that the hue signature fulfills the requirements about auto updating SDB and robustness for attackers' manipulating. This commonness can effectively expand the scope of detection when auto updating is applied to the hue signature. By the computer simulation with a real dataset, we demonstrate that our system achieves high detection performance compared with the previous scheme.

A travel route recommendation service that recommends a sequence of points of interest for tourists traveling in an unfamiliar city is a very useful tool in the field of location-based social networks. Although there are many web services and mobile applications that can help tourists to plan their trips by providing information about sightseeing attractions, travel route recommendation services are still not widely applied. One reason could be that most of the previous studies that addressed this task were based on the orienteering problem model, which mainly focuses on the estimation of a user-location relation (for example, a user preference). This assumes that a user receives a reward by visiting a point of interest and the travel route is recommended by maximizing the total rewards from visiting those locations. However, a location-location relation, which we introduce as a transition pattern in this paper, implies useful information such as visiting order and can help to improve the quality of travel route recommendations. To this end, we propose a travel route recommendation method by combining location and transition knowledge, which assigns rewards for both locations and transitions.

In this paper, we consider the clustering problem of independent general subspaces. That is, with given data points lay near or on the union of independent low-dimensional linear subspaces, we aim to recover the subspaces and assign the corresponding label to each data point. To settle this problem, we take advantages of both greedy strategy and energy minimization strategy to propose a simple yet effective algorithm based on the assumption that an m-branched (i.e., perfect m-ary) tree which is constructed by collecting m-nearest neighbor points in each node has a high probability of containing the near-exact subspace. Specifically, at first, subspace candidates are enumerated by multiple m-branched trees. Each tree starts with a data point and grows by collecting nearest neighbors in the breadth-first search order. Then, subspace proposals are further selected from the enumeration to initialize the energy minimization algorithm. Eventually, both the proposals and the labeling result are finalized by iterative re-estimation and labeling. Experiments with both synthetic and real-world data show that the proposed method can outperform state-of-the-art methods and is practical in real application.

Stance detection in twitter aims at mining user stances expressed in a tweet towards a single or multiple target entities. Detecting and analyzing user stances from massive opinion-oriented twitter posts provide enormous opportunities to journalists, governments, companies, and other organizations. Most of the prior studies have explored the traditional deep learning models, e.g., long short-term memory (LSTM) and gated recurrent unit (GRU) for detecting stance in tweets. However, compared to these traditional approaches, recently proposed densely connected bidirectional LSTM and nested LSTMs architectures effectively address the vanishing-gradient and overfitting problems as well as dealing with long-term dependencies. In this paper, we propose a neural network model that adopts the strengths of these two LSTM variants to learn better long-term dependencies, where each module coupled with an attention mechanism that amplifies the contribution of important elements in the final representation. We also employ a multi-kernel convolution on top of them to extract the higher-level tweet representations. Results of extensive experiments on single and multi-target benchmark stance detection datasets show that our proposed method achieves substantial improvement over the current state-of-the-art deep learning based methods.

In this paper we design a fast fabric defect detection framework (Fast-DDF) based on gray histogram back-projection, which adopts end to end multi-convoluted network model to realize defect classification. First, the back-projection image is established through the gray histogram on fabric image, and the closing operation and adaptive threshold segmentation method are performed to screen the impurity information and extract the defect regions. Then, the defect images segmented by the Fast-DDF are marked and normalized into the multi-layer convolutional neural network for training. Finally, in order to solve the problem of difficult adjustment of network model parameters and long training time, some strategies such as batch normalization of samples and network fine tuning are proposed. The experimental results on the TILDA database show that our method can deal with various defect types of textile fabrics. The average detection accuracy with a higher rate of 96.12% in the database of five different defects, and the single image detection speed only needs 0.72s.

This paper proposes a novel interaction technique to transfer data across various types of digital devices in uniform a manner and to allow specifying what kind of data should be sent. In our framework, when users tap multiple devices rhythmically, data corresponding to the rhythm (transfer type) are transferred from a device tapped in the first tap (source device) to the other (target device). It is easy to operate, applicable to a wide range of devices, and extensible in a sense that we can adopt new transfer types by adding new rhythms. Through a subjective evaluation and a simulation, we had a prospect that our approach would be feasible. We also discuss suggestions and limitation to implement the technique.

According to the official TDOAS 2009~2017 survey, the penetration rate of social media in Taiwan has reached a record 96.8%, while the Internet access rate is as high as 99.7%. However, people using government online services access to relevant information has continued to decline over the years, from 50.8% in 2009 to 35.4% in 2017. At the same time, the proportion of e-transaction users has also dropped simultaneously from 30.3% to 27.7%. In particular, only 1.1% of them are interested in government online forums, while the remaining 97.2% are more willing to engage in social media as a source of personal reference. The study aims to explore why are users not interested in accessing e-government services? Are they affected by the popularity of social networking applications? What are the key factors for users to continue to use e-government service? The research framework was adapted from expectation confirmation theory and model (ECT/ECM), technology acceptance model (TAM) with trust theories, in validating attitude measures for a better understanding of continuance intention of using e-government service. In terms of measurement, the assessment used the structural equation modeling method (SEM) to explore the views and preferences of 400 college students on e-government service. The study results identified that perceived usefulness not only plays a full mediating role, it is expected to be the most important ex-post factor influencing user's intention to continue using e-government service. It also clarifies that the intent to continue to use e-government services is not related to use any alternative means such as social media application.

Spoofing attacks are one of the biggest concerns for most biometric recognition systems. This will be also the case with silhouette-based gait recognition in the near future. So far, gait recognition has been fortunately out of the scope of spoofing attacks. However, it is becoming a real threat with the rapid growth and spread of deep neural network-based multimedia generation techniques, which will allow attackers to generate a fake video of gait silhouettes resembling a target person's walking motion. We refer to such computer-generated fake silhouettes as gait silhouette clones (GSCs). To deal with the future threat caused by GSCs, in this paper, we propose a supervised method for discriminating GSCs from genuine gait silhouettes (GGSs) that are observed from actual walking people. For training a good discriminator, it is important to collect training datasets of both GGSs and GSCs which do not differ from each other in any aspect other than genuineness. To this end, we propose to generate a training set of GSCs from GGSs by transforming them using multiple autoencoders. The generated GSCs are used together with their original GGSs for training the discriminator. In our experiments, the proposed method achieved the recognition accuracy of up to 94% for several test datasets, which demonstrates the effectiveness and the generality of the proposed method.

Consider selecting points on a contour in the x-y plane. In shape analysis, this is frequently referred to as contour sampling. It is important to select the points such that they effectively represent the shape of the contour. Generally, the stroke order and number of strokes are informative for that purpose. Several effective methods exist for sampling contours drawn with a certain stroke order and number of strokes, such as the English alphabet or Arabic figures. However, many contours entail an uncertain stroke order and number of strokes, such as pictures of symbols, and little research has focused on methods for sampling such contours. This is because selecting the points in this case typically requires a large computational cost to check all the possible choices. In this paper, we present a sampling method that is useful regardless of whether the contours are drawn with a certain stroke order and number of strokes or not. Our sampling method thereby expands the application possibilities of contour processing. We formulate contour sampling as a discrete optimization problem that can be solved using a type of direct search. Based on a geometric graph whose vertices are the points and whose edges form rectangles, we construct an effective objective function for the problem. Using different shape datasets, we demonstrate that our sampling method is effective with respect to shape representation and retrieval.

This paper demonstrates latent word recurrent neural network language models (LW-RNN-LMs) for enhancing automatic speech recognition (ASR). LW-RNN-LMs are constructed so as to pick up advantages in both recurrent neural network language models (RNN-LMs) and latent word language models (LW-LMs). The RNN-LMs can capture long-range context information and offer strong performance, and the LW-LMs are robust for out-of-domain tasks based on the latent word space modeling. However, the RNN-LMs cannot explicitly capture hidden relationships behind observed words since a concept of a latent variable space is not present. In addition, the LW-LMs cannot take into account long-range relationships between latent words. Our idea is to combine RNN-LM and LW-LM so as to compensate individual disadvantages. The LW-RNN-LMs can support both a latent variable space modeling as well as LW-LMs and a long-range relationship modeling as well as RNN-LMs at the same time. From the viewpoint of RNN-LMs, LW-RNN-LM can be considered as a soft class RNN-LM with a vast latent variable space. In contrast, from the viewpoint of LW-LMs, LW-RNN-LM can be considered as an LW-LM that uses the RNN structure for latent variable modeling instead of an n-gram structure. This paper also details a parameter inference method and two kinds of implementation methods, an n-gram approximation and a Viterbi approximation, for introducing the LW-LM to ASR. Our experiments show effectiveness of LW-RNN-LMs on a perplexity evaluation for the Penn Treebank corpus and an ASR evaluation for Japanese spontaneous speech tasks.

Recently, social relation analysis receives an increasing amount of attention from text to image data. However, social relation analysis from video is an important problem, which is lacking in the current literature. There are still some challenges: 1) it is hard to learn a satisfactory mapping function from low-level pixels to high-level social relation space; 2) how to efficiently select the most relevant information from noisy and unsegmented video. In this paper, we present an Attentive Sequences Recurrent Network model, called ASRN, to deal with the above challenges. First, in order to explore multiple clues, we design a Multiple Feature Attention (MFA) mechanism to fuse multiple visual features (i.e. image, motion, body, and face). Through this manner, we can generate an appropriate mapping function from low-level video pixels to high-level social relation space. Second, we design a sequence recurrent network based on Global and Local Attention (GLA) mechanism. Specially, an attention mechanism is used in GLA to integrate global feature with local sequence feature to select more relevant sequences for the recognition task. Therefore, the GLA module can better deal with noisy and unsegmented video. At last, extensive experiments on the SRIV dataset demonstrate the performance of our ASRN model.

The aim of this paper is to propose effective attentional regions for fine-grained visual recognition. Based on the Spatial Transformers' capability of spatial manipulation within networks, we propose an extension model, the Attention-Guided Spatial Transformer Networks (AG-STNs). This model can guide the Spatial Transformers with hard-coded attentional regions at first. Then such guidance can be turned off, and the network model will adjust the region learning in terms of the location and scale. Such adjustment is conditioned to the classification loss so that it is actually optimized for better recognition results. With this model, we are able to successfully capture detailed attentional information. Also, the AG-STNs are able to capture attentional information in multiple levels, and different levels of attentional information are complementary to each other in our experiments. A fusion of them brings better results.

In this paper, we study domain adaptation of semantic role classification. Most systems utilize the supervised method for semantic role classification. But, these methods often suffer severe performance drops on out-of-domain test data. The reason for the performance drops is that there are giant feature differences between source and target domain. This paper proposes a framework called Adversarial Domain Adaption Network (ADAN) to relieve domain adaption of semantic role classification. The idea behind our method is that the proposed framework can derive domain-invariant features via adversarial learning and narrow down the gap between source and target feature space. To evaluate our method, we conduct experiments on English portion in the CoNLL 2009 shared task. Experimental results show that our method can largely reduce the performance drop on out-of-domain test data.

Linear complexity and the k-error linear complexity of periodic sequences are the important security indices of stream cipher systems. This paper focuses on the distribution of p-error linear complexity of p-ary sequences with period pⁿ. For p-ary sequences of period pⁿ with linear complexity pⁿ-p+1, n≥1, we present all possible values of the p-error linear complexity, and derive the exact formulas to count the number of the sequences with any given p-error linear complexity.

A wireless sensor network consists of spatially distributed devices using sensors to monitor physical and environmental conditions. Key infection is a key distribution protocol for wireless sensor networks with a partially present adversary; a sensor node wishing to communicate secretly with other nodes simply sends a symmetric encryption key in the clear. The partially present adversary can eavesdrop on only a small fraction of the keys. Secrecy amplification is a post-deployment strategy to improve the security of key infection by combining multiple keys propagated along different paths. The previous mathematical analysis of secrecy amplification assumes that sensor nodes always transmit packets at the maximum strength. We provide a mathematical analysis of secrecy amplification where nodes adjust their transmission power adaptively (a.k.a. whispering mode).

SPHINCS⁺, an updated version of SPHINCS, is a post-quantum hash-based signature scheme submitted to the NIST post-quantum cryptography standardization project. To evaluate its performance, SPHINCS⁺ gives the theoretical number of function calls and the actual runtime of a reference implementation. We show that the theoretical number of function calls for SPHINCS⁺ verification is inconsistent with the runtime and then present the correct number of function calls.

This paper investigates the spectrum allocation problem. Under the current spectrum management mode, large amount of spectrum resource is wasted due to uncertainty of user's demand. To reduce the impact of uncertainty, a presale mechanism is designed based on spectrum pool. In this mechanism, the spectrum manager provides spectrum resource at a favorable price for presale aiming at sharing with user the risk caused by uncertainty of demand. Because of the hierarchical characteristic, we build a spectrum market Stackelberg game, in which the manager acts as leader and user as follower. Then proof of the uniqueness and optimality of Stackelberg Equilibrium is given. Simulation results show the presale mechanism can promote profits for both sides and reduce temporary scheduling.

A Service Function Chain (SFC) is an ordered sequence of virtual network functions (VNFs) to provide network service. Most existing SFC orchestration schemes, however, cannot optimize the resources allocation while guaranteeing the service delay constraint. To fulfill this goal, we propose a Layered Graph based SFC Orchestration Scheme (LGOS). LGOS converts both the cost of resource and the related delay into the link weights in the layered graph, which helps abstract the SFC orchestration problem as a shortest path problem. Then a simulated annealing based batch processing algorithm is designed for SFC requests set. Through extensive evaluations, we demonstrated that our scheme can reduce the end-to-end delay and the operational expenditure by 21.6% and 13.7% at least, and the acceptance ratio of requests set can be improved by 22.3%, compared with other algorithms.

In-class species classification based on animal sounds is a highly challenging task even with the latest deep learning technique applied. The difficulty of distinguishing the species is further compounded when the number of species is large within the same class. This paper presents a novel approach for fine categorization of animal species based on their sounds by using pre-trained CNNs and a new self-attention module well-suited for acoustic signals The proposed method is shown effective as it achieves average species accuracy of 98.37% and the minimum species accuracy of 94.38%, the highest among the competing baselines, which include CNN's without self-attention and CNN's with CBAM, FAM, and CFAM but without pre-training.

Detecting community structures and analyzing temporal evolution in dynamic networks are challenging tasks to explore the inherent characteristics of the complex networks. In this paper, we propose a semi-supervised evolutionary clustering model based on symmetric nonnegative matrix factorization to detect communities in dynamic networks, named sEC-SNMF. We use the results of community partition at the previous time step as the priori information to modify the current network topology, then smooth-out the evolution of the communities and reduce the impact of noise. Furthermore, we introduce a community transition probability matrix to track and analyze the temporal evolutions. Different from previous algorithms, our approach does not need to know the number of communities in advance and can deal with the situation in which the number of communities and nodes varies over time. Extensive experiments on synthetic datasets demonstrate that the proposed method is competitive and has a superior performance.

Convolutional Neural Networks (CNNs) usually have millions or even billions of parameters, which make them hard to be deployed into mobile devices. In this work, we present a novel filter-level pruning method to alleviate this issue. More concretely, we first construct an undirected fully connected graph to represent a pre-trained CNN model. Then, we employ the spectral clustering algorithm to divide the graph into some subgraphs, which is equivalent to clustering the similar filters of the CNN into the same groups. After gaining the grouping relationships among the filters, we finally keep one filter for one group and retrain the pruned model. Compared with previous pruning methods that identify the redundant filters by heuristic ways, the proposed method can select the pruning candidates more reasonably and precisely. Experimental results also show that our proposed pruning method has significant improvements over the state-of-the-arts.

Hand-dorsa vein recognition is solved based on the convolutional activations of the pre-trained deep convolutional neural network (DCNN). In specific, a novel task-specific cross-convolutional-layer pooling is proposed to obtain the more representative and discriminative feature representation. Rigorous experiments on the self-established database achieves the state-of-the-art recognition result, which demonstrates the effectiveness of the proposed model.

For cross-corpus speech emotion recognition (SER), how to obtain effective feature representation for the discrepancy elimination of feature distributions between source and target domains is a crucial issue. In this paper, we propose a Target-adapted Subspace Learning (TaSL) method for cross-corpus SER. The TaSL method trys to find a projection subspace, where the feature regress the label more accurately and the gap of feature distributions in target and source domains is bridged effectively. Then, in order to obtain more optimal projection matrix, ℓ₁ norm and ℓ_2,1 norm penalty terms are added to different regularization terms, respectively. Finally, we conduct extensive experiments on three public corpuses, EmoDB, eNTERFACE and AFEW 4.0. The experimental results show that our proposed method can achieve better performance compared with the state-of-the-art methods in the cross-corpus SER tasks.

We develop a deep neural network (DNN) for detecting driver drowsiness in videos. The proposed DNN model that receives driver's faces extracted from video frames as inputs consists of three components - a convolutional neural network (CNN), a convolutional control gate-based recurrent neural network (ConvCGRNN), and a voting layer. The CNN is to learn facial representations from global faces which are then fed to the ConvCGRNN to learn their temporal dependencies. The voting layer works like an ensemble of many sub-classifiers to predict drowsiness state. Experimental results on the NTHU-DDD dataset show that our model not only achieve a competitive accuracy of 84.81% without any post-processing but it can work in real-time with a high speed of about 100 fps.

A non-photorealistic rendering method has been proposed for generating oil-film-like images from photographic images by bilateral infra-envelope filter. The conventional method has a disadvantage that it takes much time to process. We propose a method for generating oil-film-like images that can be processed faster than the conventional method. The proposed method uses an iterative process with upper and lower smoothing filters. To verify the effectiveness of the proposed method, we conduct experiments using Lenna image. As a result of the experiments, we show that the proposed method can process faster than the conventional method.

Convolutional neural networks (CNNS) have a strong ability to understand and judge images. However, the enormous parameters and computation of CNNS have limited its application in resource-limited devices. In this letter, we used the idea of parameter sharing and dense connection to compress the parameters in the convolution kernel channel direction, thus greatly reducing the number of model parameters. On this basis, we designed Shared and Dense Channel-wise Convolutional Networks (SDChannelNets), mainly composed of Depth-wise Separable SD-Channel-wise Convolution layer. The advantage of SDChannelNets is that the number of model parameters is greatly reduced without or with little loss of accuracy. We also introduced a hyperparameter that can effectively balance the number of parameters and the accuracy of a model. We evaluated the model proposed by us through two popular image recognition tasks (CIFAR-10 and CIFAR-100). The results showed that SDChannelNets had similar accuracy to other CNNs, but the number of parameters was greatly reduced.