Analysis of large-scale sequential data has been one of the most crucial tasks in areas such as bioinformatics, text, and audio mining. Existing string kernels, however, either (i) rely on local features of short substructures in the string, which hardly capture long discriminative patterns, (ii) sum over too many substructures, such as all possible subsequences, which leads to diagonal dominance of the kernel matrix, or (iii) rely on non-positive-definite similarity measures derived from the edit distance. Furthermore, while there have been works addressing the computational challenge with respect to the length of string, most of them still experience quadratic complexity in terms of the number of training samples when used in a kernel-based classifier. In this paper, we present a new class of global string kernels that aims to (i) discover global properties hidden in the strings through global alignments, (ii) maintain positive-definiteness of the kernel, without introducing a diagonal dominant kernel matrix, and (iii) have a training cost linear with respect to not only the length of the string but also the number of training string samples. To this end, the proposed kernels are explicitly defined through a series of different random feature maps, each corresponding to a distribution of random strings. We show that kernels defined this way are always positive-definite, and exhibit computational benefits as they always produce \emph{Random String Embeddings (RSE)} that can be directly used in any linear classification models. Our extensive experiments on nine benchmark datasets corroborate that RSE achieves better or comparable accuracy in comparison to state-of-the-art baselines, especially with the strings of longer lengths. In addition, we empirically show that RSE scales linearly with the increase of the number and the length of string.
Effectively capturing graph node sequences in the form of vector embeddings is critical to many applications. We achieve this by (i) first learning vector embeddings of single graph nodes and (ii) then composing them to compactly represent node sequences. Specifically, we propose SENSE-S (Semantically Enhanced Node Sequence Embedding - for Single nodes), a skip-gram based novel embedding mechanism, for single graph nodes that co-learns graph structure as well as their textual descriptions. We demonstrate that SENSE-S vectors increase the accuracy of multi-label classification tasks by up to 50% and link-prediction tasks by up to 78% under a variety of scenarios using real datasets. Based on SENSE-S, we next propose generic SENSE to compute composite vectors that represent a sequence of nodes, where preserving the node order is important. We prove that this approach is efficient in embedding node sequences, and our experiments on real data confirm its high accuracy in node order decoding.
In distributed software-defined networks (SDN), multiple physical SDN controllers, each managing a network domain, are implemented to balance centralised control, scalability, and reliability requirements. In such networking paradigms, controllers synchronize with each other, in attempts to maintain a logically centralised network view. Despite the presence of various design proposals for distributed SDN controller architectures, most existing works only aim at eliminating anomalies arising from the inconsistencies in different controllers' network views. However, the performance aspect of controller synchronization designs with respect to given SDN applications are generally missing. To fill this gap, we formulate the controller synchronization problem as a Markov decision process (MDP) and apply reinforcement learning techniques combined with deep neural networks (DNNs) to train a smart, scalable, and fine-grained controller synchronization policy, called the Multi-Armed Cooperative Synchronization (MACS), whose goal is to maximise the performance enhancements brought by controller synchronizations. Evaluation results confirm the DNN's exceptional ability in abstracting latent patterns in the distributed SDN environment, rendering significant superiority to MACS-based synchronization policy, which are 56% and 30% performance improvements over ONOS and greedy SDN controller synchronization heuristics.
Person re-identification (ReID) aims to match people across multiple non-overlapping video cameras deployed at different locations. To address this challenging problem, many metric learning approaches have been proposed, among which triplet loss is one of the state-of-the-arts. In this work, we explore the margin between positive and negative pairs of triplets and prove that large margin is beneficial. In particular, we propose a novel multi-stage training strategy which learns incremental triplet margin and improves triplet loss effectively. Multiple levels of feature maps are exploited to make the learned features more discriminative. Besides, we introduce global hard identity searching method to sample hard identities when generating a training batch. Extensive experiments on Market-1501, CUHK03, and DukeMTMCreID show that our approach yields a performance boost and outperforms most existing state-of-the-art methods.
Using FPGAs to accelerate ConvNets has attracted significant attention in recent years. However, FPGA accelerator design has not leveraged the latest progress of ConvNets. As a result, the key application characteristics such as frames-per-second (FPS) are ignored in favor of simply counting GOPs, and results on accuracy, which is critical to application success, are often not even reported. In this work, we adopt an algorithm-hardware co-design approach to develop a ConvNet accelerator called Synetgy and a novel ConvNet model called DiracNet. Both the accelerator and ConvNet are tailored to FPGA requirements. DiractNet, as the name suggests, is a ConvNet with only 1x1 convolutions while spatial convolutions are replaced by more efficient shift operations. DiracNet achieves competitive accuracy on ImageNet (89.0% top-5), but with 48x fewer parameters and 65x fewer OPs than VGG16. We further quantize DiracNet's weights to 1-bit and activations to 4-bits, with less than 1% accuracy loss. These quantizations exploit well the nature of FPGA hardware. In short, DiracNet's small model size, low computational OP count, ultra-low precision and simplified operators allow us to co-design a highly customized computing unit for an FPGA. We implement the computing units for DiracNet on an Ultra96 SoC system through high-level synthesis. The implementation only took 2 people 1 month to complete. Our accelerator's final top-5 accuracy of 88.3% on ImageNet, is higher than all the previously reported embedded FPGA accelerators. In addition, the accelerator reaches an inference speed of 72.8 FPS on the ImageNet classification task, surpassing prior works with similar accuracy by at least 12.8x.
Cumulative local muscle fatigue may lead to potential musculoskeletal disorder (MSD) risks {\color{red}, and subject-specific muscle fatigability needs to be considered to reduce potential MSD risks.} This study was conducted to determine local muscle fatigue rate at shoulder joint level based on an exponential function derived from a muscle fatigue model. Forty male subjects participated in a fatiguing operation under a static posture with a range of relative force levels (14% - 33%). Remaining maximum muscle strengths were measured after different fatiguing sessions. The time course of strength decline was fitted to the exponential function. Subject-specific fatigue rates of shoulder joint moment strength were determined. Good correspondence ($R^2>0.8$) was found in the regression of the majority (35 out of 40 subjects). Substantial inter-individual variability in fatigue rate was found and discussed.
During the conceptual and prototype design stage of an industrial product, it is crucial to take assembly/disassembly and maintenance operations in advance. A well-designed system should enable relatively easy access of operating manipulators in the constrained environment and reduce musculoskeletal disorder risks for those manual handling operations. Trajectory planning comes up as an important issue for those assembly and maintenance operations under a constrained environment, since it determines the accessibility and the other ergonomics issues, such as muscle effort and its related fatigue. In this paper, a customer-oriented interactive approach is proposed to partially solve ergonomic related issues encountered during the design stage under a constrained system for the operator's convenience. Based on a single objective optimization method, trajectory planning for different operators could be generated automatically. Meanwhile, a motion capture based method assists the operator to guide the trajectory planning interactively when either a local minimum is encountered within the single objective optimization or the operator prefers guiding the virtual human manually. Besides that, a physical engine is integrated into this approach to provide physically realistic simulation in real time manner, so that collision free path and related dynamic information could be computed to determine further muscle fatigue and accessibility of a product design
Human muscle fatigue is considered to be one of the main reasons for Musculoskeletal Disorder (MSD). Recent models have been introduced to define muscle fatigue for static postures. However, the main drawbacks of these models are that the dynamic effect of the human and the external load are not taken into account. In this paper, each human joint is assumed to be controlled by two muscle groups to generate motions such as push/pull. The joint torques are computed using Lagrange's formulation to evaluate the dynamic factors of the muscle fatigue model. An experiment is defined to validate this assumption and the result for one person confirms its feasibility. The evaluation of this model can predict the fatigue and MSD risk in industry production quickly.
This paper presents our work on relationship of evaluation results between virtual environment (VE) and realistic environment (RE) for assembling tasks. Evaluation results consist of subjective results (BPD and RPE) and objective results (posture and physical performance). Same tasks were performed with same experimental configurations and evaluation results were measured in RE and VE respectively. Then these evaluation results were compared. Slight difference of posture between VE and RE was found but not great difference of effect on people according to conventional ergonomics posture assessment method. Correlation of BPD and performance results between VE and RE are found by linear regression method. Moreover, results of BPD, physical performance, and RPE in VE are higher than that in RE with significant difference. Furthermore, these results indicates that subjects feel more discomfort and fatigue in VE than RE because of additional effort required in VE.
In ergonomics and biomechanics, muscle fatigue models based on maximum endurance time (MET) models are often used to integrate fatigue effect into ergonomic and biomechanical application. However, due to the empirical principle of those MET models, the disadvantages of this method are: 1) the MET models cannot reveal the muscle physiology background very well; 2) there is no general formation for those MET models to predict MET. In this paper, a theoretical MET model is extended from a simple muscle fatigue model with consideration of the external load and maximum voluntary contraction in passive static exertion cases. The universal availability of the extended MET model is analyzed in comparison to 24 existing empirical MET models. Using mathematical regression method, 21 of the 24 MET models have intraclass correlations over 0.9, which means the extended MET model could replace the existing MET models in a general and computationally efficient way. In addition, an important parameter, fatigability (or fatigue resistance) of different muscle groups, could be calculated via the mathematical regression approach. Its mean value and its standard deviation are useful for predicting MET values of a given population during static operations. The possible reasons influencing the fatigue resistance were classified and discussed, and it is still a very challenging work to find out the quantitative relationship between the fatigue resistance and the influencing factors.