Search

Large language model (LLM) has demonstrated significant application potential in the medical field. However, evaluating the performance of LLM in medical scenarios poses a challenge. Existing medical benchmarks, predominantly in the form of multiple-choice questions, struggle to comprehensively and accurately assess LLM's performance in pediatric domains. To address this issue, PeMeBench, the first Chinese pediatric question-answering benchmark, was proposed. Leveraging a dual-perspective evaluation dimensions and referencing diagnostic and treatment guidelines from 10 pediatric disease systems, PeMeBench meticulously categorized pediatric medical question-answering tasks into five subdomains: disease knowledge, treatment plans, medication dosages, disease prevention, and pharmacological effects. It comprised over 10 000 open-ended question-answering items and introduced a multi-grained automated evaluation scheme that integrated entity retrieval with the detection of hallucinated sentences. This approach aimed to provide a comprehensive and precise assessment of LLM's performance in pediatric healthcare, delving into their potential limitations and laying a solid foundation for enhancing the intelligence level of medical services.

Large variations across images make image classification a challenging task; limited training data further increases its difficulty. Genetic programming (GP) has been considerably applied to image classification. However, most GP methods tend to directly evolve a single classifier or depend on a predefined classification algorithm, which typically does not lead to ideal generalization performance when only a few training instances are available. Applying ensemble learning to classification often outperforms employing a single classifier. However, single-tree representation (each individual contains a single tree) is widely employed in GP. Training multiple diverse and accurate base learners/classifiers based on single-tree GP is challenging. Therefore, this article proposes a new ensemble construction method based on multi-tree GP (each individual contains multiple trees) for image classification. A single individual forms an ensemble, and its multiple trees constitute base learners. To find the best individual in which multiple trees are diverse and effectively cooperate, i.e., the nth tree can correct the errors of the previous n-1 trees, the new method assigns different weights to multiple trees using the idea of AdaBoost and performs classification via weighted majority voting. Furthermore, a new tree representation is developed to evolve diverse and accurate base learners that extract useful features and conduct classification simultaneously. The new approach achieves significantly better performance than almost all benchmark methods on eight datasets. Additional analyses highlight the effectiveness of the new ensembles and tree representation, demonstrating the potential for providing valuable interpretability in ensemble trees. [ABSTRACT FROM AUTHOR]

Meta-heuristic algorithms, especially evolutionary algorithms, have been frequently used to find near optimal solutions to combinatorial optimization problems. The evaluation of such algorithms is often conducted through comparisons with other algorithms on a set of benchmark problems. However, even if one algorithm is the best among all those compared, it still has difficulties in determining the true quality of the solutions found because the true optima are unknown, especially in dynamic environments. It would be desirable to evaluate algorithms not only relatively through comparisons with others, but also in absolute terms by estimating their quality compared to the true global optima. Unfortunately, true global optima are normally unknown or hard to find since the problems being addressed are NP-hard. In this paper, instead of using true global optima, lower bounds are derived to carry out an objective evaluation of the solution quality. In particular, the first approach capable of deriving a lower bound for dynamic capacitated arc routing problem (DCARP) instances is proposed, and two optimization algorithms for DCARP are evaluated based on such a lower bound approach. An effective graph pruning strategy is introduced to reduce the time complexity of our proposed approach. Our experiments demonstrate that our approach provides tight lower bounds for small DCARP instances. Two optimization algorithms are evaluated on a set of DCARP instances through the derived lower bounds in our experimental studies, and the results reveal that the algorithms still have room for improvement for large complex instances. [ABSTRACT FROM AUTHOR]

Recent advances in tumor molecular subtyping have revolutionized precision oncology, offering novel avenues for patient-specific treatment strategies. However, a comprehensive and independent comparison of these subtyping methodologies remains unexplored. This study introduces 'Themis' (Tumor HEterogeneity analysis on Molecular subtypIng System), an evaluation platform that encapsulates a few representative tumor molecular subtyping methods, including Stemness, Anoikis, Metabolism, and pathway-based classifications, utilizing 38 test datasets curated from The Cancer Genome Atlas (TCGA) and significant studies. Our self-designed quantitative analysis uncovers the relative strengths, limitations, and applicability of each method in different clinical contexts. Crucially, Themis serves as a vital tool in identifying the most appropriate subtyping methods for specific clinical scenarios. It also guides fine-tuning existing subtyping methods to achieve more accurate phenotype-associated results. To demonstrate the practical utility, we apply Themis to a breast cancer dataset, showcasing its efficacy in selecting the most suitable subtyping methods for personalized medicine in various clinical scenarios. This study bridges a crucial gap in cancer research and lays a foundation for future advancements in individualized cancer therapy and patient management. [ABSTRACT FROM AUTHOR]

Modern computer designs support composite prefetching, where multiple prefetcher components are used to target different memory access patterns. However, multiple prefetchers competing for resources can sometimes hurt performance, especially in many-core systems where cache and other resources are limited. Recent work has proposed mitigating this issue by selectively enabling and disabling prefetcher components at runtime. Formulating the problem with machine learning (ML) methods is promising, but efficient and effective solutions in terms of cost and performance are not well understood. This work studies fundamental characteristics of the composite prefetcher selection problem through the lens of ML to inform future prefetcher selection designs. We show that prefetcher decisions do not have significant temporal dependencies, that a phase-based rather than sample-based definition of ground truth yields patterns that are easier to learn, and that prefetcher selection can be formulated as a workload-agnostic problem requiring little to no training at runtime. [ABSTRACT FROM AUTHOR]

Personality determines various human daily and working behaviours. Recently, a large number of automatic personality computing approaches have been developed to predict either the apparent or self-reported personality of the subject based on non-verbal audio-visual behaviours. However, most of them suffer from complex and dataset-specific pre-processing steps and model training tricks. In the absence of a standardized benchmark with consistent experimental settings, it is not only impossible to fairly compare the real performances of these personality computing models but also makes them difficult to be reproduced. This paper presents the first reproducible audio-visual benchmark to provide a fair and consistent evaluation of eight existing personality computing models (e.g., audio, visual and audio-visual) and seven standard deep learning models on both self-reported and apparent personality recognition tasks. Building upon a set of benchmarked models, we also investigate the impact of two previously-used long-term modelling strategies for summarising short-term/frame-level predictions on personality computing results. We comprehensively investigate all benchmarked models on two publicly available datasets, ChaLearn First Impression and UDIVA self-reported personality datasets, and conclude: (i) apparent personality traits, inferred from facial behaviours by most benchmarked deep learning models, show more reliability than self-reported ones; (ii) visual models frequently achieved superior performances than audio models on personality recognition; (iii) non-verbal behaviours contribute differently in predicting different personality traits; and (iv) our reproduced personality computing models generally achieved worse performances than their original reported results. [ABSTRACT FROM AUTHOR]

Imaging of facial affects may be used to measure psychophysiological attributes of children through their adulthood for applications in education, healthcare, and entertainment, among others. Deep convolutional neural networks show promising results in classifying facial expressions of adults. However, classifier models trained with adult benchmark data are unsuitable for learning child expressions due to discrepancies in psychophysical development. Similarly, models trained with child data perform poorly in adult expression classification. We propose domain adaptation to concurrently align distributions of adult and child expressions in a shared latent space for robust classification of either domain. Furthermore, age variations in facial images are studied in age-invariant face recognition yet remain unleveraged in adult-child expression classification. We take inspiration from multiple fields and propose deep adaptive FACial Expressions fusing BEtaMix SElected Landmark Features (FACE-BE-SELF) for adult-child expression classification. For the first time in the literature, a mixture of Beta distributions is used to decompose and select facial features based on correlations with expression, domain, and identity factors. We evaluate FACE-BE-SELF using 5-fold cross validation for two pairs of adult-child data sets. Our proposed FACE-BE-SELF approach outperforms transfer learning and other baseline domain adaptation methods in aligning latent representations of adult and child expressions. [ABSTRACT FROM AUTHOR]

Previous studies have demonstrated that the response of stable yielding systems to random excitations such as earthquakes is inherently accompanied by residual deformations. Stable yielding systems with highly enhanced ductility capacity are more prone to residual deformations. Steel eccentrically braced frames (EBFs) are designed to sustain significant localized inelastic deformations in the segment of the beam referred to as the yielding link, which results in much larger link rotations when compared to inter‐story drift ratios (SDR). Recent experimental and numerical studies on the response of EBFs have demonstrated that even with moderate residual inter‐SDR (less than 0.5%), severe link residual rotations could be expected, which could render the structure difficult to repair. For replaceable yielding links, which were developed to expedite the repair and recovery process in EBFs after major seismic events, large residual link rotations will hinder the link replacement process or at times make it impossible. Therefore, mitigation of residual deformations is of greater importance in EBFs with replaceable yielding links than in conventional EBFs with a continuous beam. This paper provides a brief review of past research on residual deformations and methods for their mitigation. A low‐cost re‐centering system for use in EBFs is proposed to work in parallel with the yielding link, in order to mitigate residual deformations. The proposed system is adopted for two prototype structures designed with EBFs equipped with cast steel replaceable modular yielding links. The effectiveness of the re‐centering strategy is evaluated by means of pseudo‐dynamic hybrid simulations, as well as numerical simulations. The hybrid simulations and numerical studies both demonstrate that not only the proposed system is highly effective at mitigating the residual deformations, but also it is effective in reducing the peak deformations in EBFs. [ABSTRACT FROM AUTHOR]

Detecting vulnerable code blocks has become a highly popular topic in computer-aided design, especially with the advancement of natural language processing (NLP). Analyzing hardware description languages (HDLs), such as Verilog, involves dealing with lengthy code. This letter introduces an innovative identification of attack-vulnerable hardware by the use of opcode processing. Leveraging the advantage of architecturally defined opcodes and expressing all operations at the beginning of each code line, the word processing problem is efficiently transformed into opcode processing. This research converts a benchmark dataset into an intermediary code stack, subsequently classifying secure and fragile codes using NLP techniques. The results reveal a framework that achieves up to 94% accuracy when employing sophisticated convolutional neural networks (CNNs) architecture with extra embedding layers. Thus, it provides a means for users to quickly verify the vulnerability of their HDL code by inspecting a supervised learning model trained on the predefined vulnerabilities. It also supports the superior efficacy of opcode-based processing in Trojan detection by analyzing the outcomes derived from a model trained using the HDL dataset. [ABSTRACT FROM AUTHOR]

Due to the increasing complexity of modern integrated circuits, high-level synthesis (HLS) is becoming a key technology in hardware design. HLS uses optimizations to assist during design space exploration. However, some of them can introduce security weaknesses. We propose an approach that leverages static analysis to identify a class of weaknesses in HLS-generated code. We show that some of these weaknesses can be corrected through the automatic generation of HLS directives. We evaluate our approach by comparing the static analysis results with formal verification. Our results show that the static approach has the same accuracy as formal methods while being $3\times $ to $200\times $ faster. [ABSTRACT FROM AUTHOR]

HT has become a serious threat to the Internet of Things due to the globalization of the integrated circuit industry. To evade functional verification, HTs tend to have at least one trigger signal at the gate-level netlist with a very low transition probability. Based on this nature, previous studies use imbalanced controllability as a feature to detect HTs, assuming that signals with imbalanced controllability are always accompanied by low transition probability. However, this study has found out a way to create a new type of HT that has low transition probability but balanced controllability, against previous methods. Hence, current imbalanced controllability detectors are inadequate in this scenario. To address this limitation, we propose a probability-based detection method that uses unsupervised anomaly analysis to detect HTs. Our proposed method detects not only the proposed HT but also the 580 Trojan benchmarks on Trusthub. Experimental results show that our proposed detector outperforms other detectors, achieving an overall 100% true positive rate and 0.37% false positive rate on the 580 benchmarks. [ABSTRACT FROM AUTHOR]

This work is an original contribution consisting in the implementation and evaluation of a selective value predictor in a multicore environment, with focus on long latency arithmetical instructions, having the goal to break the dataflow bottleneck of each core, thus increasing the overall performance. The Sniper simulator was used to augment the Intel Nehalem architecture with a value predictor and to estimate the computing performance, area of integration, power consumption, energy efficiency and chip temperature for the enhanced architecture. We run simulations and study the impact of the number of values which are used for prediction for each instruction. By increasing the history length, we measured on average more than 3 % increase in performance (core speed-up), a reduction in chip temperature from 57.8 °C to 56.17 °C, and lower energy consumption in most cases compared with the baseline configuration. We also realized a comparison between the value prediction and dynamic instruction reuse techniques in equitable condition (to exploit the same value locality), where we highlight the advantages and disadvantages of each technique in the given context. [ABSTRACT FROM AUTHOR]

Driver emotion is considered an essential factor associated with driving behaviors and thus influences traffic safety. Dynamically and accurately recognizing the emotions of drivers plays an important role in road safety, especially for professional drivers, e.g., the drivers of passenger service vehicles. However, there is a lack of a benchmark to quantitatively evaluate the performance of driver emotion recognition performance, especially for various application situations. In this article, we propose an emotion recognition benchmark based on the driver emotion facial expression (DEFE) dataset, which consists of two splits: training and testing on the same set (split 1) and different sets (split 2) of drivers. These two splits correspond to various application scenarios and have diverse challenges. For the former, a driver emotion recognition network is proposed to provide a competitive baseline for the benchmark. For the latter, a novel driver representation difference minimization loss is proposed to enhance the learning of common representations for emotion recognition over different drivers. Moreover, the minimum required information for achieving a satisfactory performance is also explored on split 2. Comprehensive experiments on the DEFE dataset clearly demonstrate the superiority of the proposed methods compared to other state-of-the-art methods. An example application of applying the proposed methods and a voting mechanism to real-world data collected in a naturalistic environment reveals the strong practicality and readiness of the proposed methods. The codes and dataset splits are publicly available at https://github.com/wdy806/CDERNet/. [ABSTRACT FROM AUTHOR]

Radiological source search is a challenging task involving detection and identification of weak sources in a constantly changing radiological background. As of now, many radiological source detection algorithms have been proposed; however, their computational complexity and, hence, reliance on power intensive processing units inhibit low-power applications of radiological source search systems. In this work, we introduce the anomaly filter (AF) algorithm; a computationally light, yet effective time-series source detection algorithm based on exponential weighted moving average (EWMA) and Poisson deviance statistics. Then, we demonstrate that the proposed algorithm can be used in ensemble with other more computationally intensive source detection and identification algorithms to achieve both increased detection performance and reduced power consumption. The proposed AF algorithm and the ensemble algorithms were thoroughly benchmarked against several existing source detection and identification algorithms. The results show that the AF algorithm outperforms existing conventional source detection algorithms, and the ensemble approach improves the overall performance of existing source detection and isotope identification algorithms. Furthermore, the AF algorithm and the non-negative matrix factorization approach-based source identification (NMF-ID) algorithm were combined and implemented on a single-board microcontroller, and the power consumption was measured. This ensemble algorithm reduced the power consumption of the NMF-ID algorithm almost by a factor of 100, while improving the detection performance of the overall system.

Fingerprint recognition remains a cornerstone of biometric identification systems. However, the critical task of ridge-line frequency estimation has been surprisingly underexplored. To address this research gap, I introduce the first publicly available benchmark for fingerprint frequency estimation, a pioneering resource that includes both high- and low-quality fingerprints with meticulously labeled ground truth features: segmentation masks, orientation fields, and frequency maps. Furthermore, two novel frequency estimation methods are proposed: one that significantly enhances a well-known frequency estimation method based on traditional image processing techniques and another that, for the first time, applies deep learning to this context. Experimental results on the new benchmark demonstrate that both methods surpass existing state-of-the-art techniques, particularly in challenging low-quality fingerprint scenarios. By providing an open-source implementation and a comprehensive benchmark, this work sets a new standard for the evaluation of frequency estimation methods, fostering further research and development in this crucial area of fingerprint recognition.

Vehicle detection in unmanned aerial vehicle (UAV) images is a fundamental task in photogrammetry and remote sensing. While great success has been achieved, this task remains challenging due to two aspects: 1) the limitation of existing annotation methods in compactly enclosing targets with large perspective distortions in oblique UAV images; 2) the lack of vehicle detection datasets under oblique perspectives. To this end, we propose an oblique UAV benchmark for the precise expression and localization of distorted vehicles in urban scenarios. The benchmark consists of 1) a new parallelogramlike bounding box (PBB) annotation for compactly representing vehicles in oblique UAV images; and 2) a large-scale UAV dataset (namely PARA) for vehicle detection with PBB representation. Our PBB representation frees the angle flexibility to allow a compact depiction of vehicles under various perspective distortion, thus overcoming the inherent limits of rectangular representation [like horizontal bounding box (HBB)] used in traditional annotation methods. PARA comprises 1025 high-resolution images and 117 122 manually annotated object bounding boxes obtained from different UAV platforms. The annotated images are collected from scenarios with complex urban backgrounds and different shooting angles to reflect real-world conditions. Moreover, we compared detection algorithms based on the mainstream HBB and PBB representations on the PARA dataset and established a baseline for UAV oblique image-based vehicle detection. Experimental results validate the effectiveness of PBB representation and highlight the challenges posed by PARA.

Human-machine systems, especially those involving reinforcement learning (RL), are becoming increasingly common across application domains. Designing these systems to be effective and reliable requires a task-oriented understanding of both human learning (HL) and RL. In particular, how does the structure of a learning task affect the learning performance of humans and RL algorithms? Games and other learning environments can serve as important tools in this line of research. While a trend toward increasingly complex environments has led to improved RL capabilities, such environments are difficult to use for the dedicated study of task structure for humans and algorithms. To address this gap we present a novel learning environment called the Game of Hidden Rules (GOHR), built to support rigorous study of the impact of task structure on HL and RL. We demonstrate the GOHR’s utility for such study through example experiments where humans and learning algorithms display opposite responses in performance to tested variations in task structure.

Choosing optimal maskers for existing soundscapes to effect a desired perceptual change via soundscape augmentation is non-trivial due to extensive varieties of maskers and a dearth of benchmark datasets with which to compare and develop soundscape augmentation models. To address this problem, we make publicly available the ARAUS (Affective Responses to Augmented Urban Soundscapes) dataset, which comprises a five-fold cross-validation set and independent test set totaling 25,440 unique subjective perceptual responses to augmented soundscapes presented as audio-visual stimuli. Each augmented soundscape is made by digitally adding “maskers” (bird, water, wind, traffic, construction, or silence) to urban soundscape recordings at fixed soundscape-to-masker ratios. Responses were then collected by asking participants to rate how pleasant, annoying, eventful, uneventful, vibrant, monotonous, chaotic, calm, and appropriate each augmented soundscape was, in accordance with ISO/TS 12913-2:2018. Participants also provided relevant demographic information and completed standard psychological questionnaires. We perform exploratory and statistical analysis of the responses obtained to verify internal consistency and agreement with known results in the literature. Finally, we demonstrate the benchmarking capability of the dataset by training and comparing four baseline models for urban soundscape pleasantness: a low-parameter regression model, a high-parameter convolutional neural network, and two attention-based networks in the literature. [ABSTRACT FROM AUTHOR]

The emerging nonvolatile memories (NVMs), such as spin transfer torque random access memory (STT-RAM) and racetrack memory (RTM), offer a promising solution to satisfy the memory and performance requirements of modern applications. Compared to the commonly utilized volatile static random-access memories (SRAMs), the NVMs provide better capacity and energy efficiency. However, many of these NVMs are still in the development phases and require proper evaluation in order to evaluate the impact of their use at the system level. Therefore, there is a need to design functional- and cycleaccurate simulators/emulators to evaluate the performance of these memory technologies. To this end, this work focuses on implementing a RISC-V-based emulation platform for evaluating NVMs. The proposed framework provides interfaces to integrate various types of NVMs, with RTMs and STT-RAMs used as test cases. The efficacy of the framework is evaluated by executing benchmark applications. [ABSTRACT FROM AUTHOR]

Memristor-based logic-in-memory (LiM) has become popular as a means to overcome the von Neumann bottleneck in traditional data-intensive computing. Recently, the memristor-aided logic (MAGIC) design style has gained immense traction for LiM due to its simplicity. However, understanding the energy distribution during the design of logic operations within the memristive memory is crucial in assessing such an implementation’s significance. The current energy estimation methods rely on coarse-grained techniques, which underestimate the energy consumption of MAGIC-styled operations performed on a memristor crossbar. To address this issue, we analyze the energy breakdown in MAGIC operations and propose a solution that utilizes mapping from the SIMPLER MAGIC tool to achieve accurate energy estimation through SPICE simulations. In contrast to existing research that primarily focuses on optimizing execution energy, our findings reveal that the memristor’s initialization energy in the MAGIC design style is, on average, $68\times $ higher. We demonstrate that this initialization energy significantly dominates the overall energy consumption. By highlighting this aspect, we aim to redirect the attention of designers toward developing algorithms and strategies that prioritize optimizations in initializations rather than execution for more effective energy savings. [ABSTRACT FROM AUTHOR]

Previous studies have underlined the importance of establishing a database of high‐fidelity benchmark experimental test results under random loading histories such as earthquake excitations for both existing structural systems as well as newly proposed ones. Such experimental results are useful for understanding the limitations of previously developed numerical models, proposing new numerical models for more recently developed structural systems, better assessing the ultra‐low cycle fatigue (ULCF) life of energy dissipative components, and improving the loading protocols used for element assessments or developments of numerical model. This paper presents experimental results from pseudo‐dynamic hybrid simulations on steel eccentrically braced frames (EBF) equipped with novel cast steel replaceable modular yielding links. The first set of experiments are carried out on a four‐story EBF, with the first‐floor yielding link physically tested in the laboratory. The second set of experiments are performed on a two‐story EBF where the second‐floor link is physically tested in the laboratory, while considering axial loads due to the imbalanced distribution of the seismic weight in the building. Each building structure is subjected to three Maximum Credible Event (MCE) level earthquakes. A framework for performing hybrid simulations on EBFs is proposed where the response of the yielding link is physically captured in a single degree of freedom control system, with or without axial loads. A substructuring strategy is proposed in conjunction with a modeling approach in the integration module. The results confirm the effectiveness of the hybrid simulation framework, which can be adopted in similar studies. Two numerical models are proposed for capturing the response of cast steel links using a simplified and a more advanced approach. The models are first calibrated using incremental reversed‐cyclic experimental test results, and then refined using the hybrid simulation test results. The effects of this calibration improvement are quantified through critically comparing the response prediction for different response parameters before and after improving the calibration with the experimental results. The numerical model which captures the physical mechanism of the cast steel links provides a more accurate response prediction compared to the simplified numerical model using a phenomenological truss with calibrated force‐deformation. The remaining ULCF life of the tested yielding links is experimentally investigated after having sustained three MCE level earthquakes, which indicates a large reserve ductility capacity which is more than twice the required plastic rotation capacity for cast steel links after seismic events. [ABSTRACT FROM AUTHOR]

Traditional solvers for tackling combinatorial optimization (CO) problems are usually designed by human experts. Recently, there has been a surge of interest in utilizing deep learning, especially deep reinforcement learning, to automatically learn effective solvers for CO. The resultant new paradigm is termed neural combinatorial optimization (NCO). However, the advantages and disadvantages of NCO relative to other approaches have not been empirically or theoretically well studied. This work presents a comprehensive comparative study of NCO solvers and alternative solvers. Specifically, taking the traveling salesman problem as the testbed problem, the performance of the solvers is assessed in five aspects, i.e., effectiveness, efficiency, stability, scalability, and generalization ability. Our results show that the solvers learned by NCO approaches, in general, still fall short of traditional solvers in nearly all these aspects. A potential benefit of NCO solvers would be their superior time and energy efficiency for small-size problem instances when sufficient training instances are available. Hopefully, this work would help with a better understanding of the strengths and weaknesses of NCO and provide a comprehensive evaluation protocol for further benchmarking NCO approaches in comparison to other approaches. [ABSTRACT FROM AUTHOR]

Cooperative co-evolution (CC) is an evolutionary algorithm that adopts the divide-and-conquer strategy to solve large-scale optimization problems. It is difficult for CC to specify a suitable subpopulation size to solve different subproblems. The population diversity may be insufficient to search for the global optimum during subpopulations' evolution. In this paper, an adaptive method for enhancing population diversity is embedded in a contribution-based CC. In CC, there are two kinds of subpopulation: the convergent or stagnant subpopulations and the non-convergent and non-stagnant subpopulations. A method is proposed in the paper to evaluate the convergent and stagnant subpopulations' contributions to improving the best overall objective value, which is different from the contribution evaluation on the non-convergent and non-stagnant subpopulations. In each co-evolutionary cycle, the new CC adaptively determines to select a subpopulation, which can make a greater contribution to improving the best overall objective value, between the above two kinds of subpopulation to undergo evolution. When a convergent or stagnant subpopulation is selected to undergo evolution, the subpopulation is re-diversified to enhance its global search capability. Our experimental results and analysis suggest that the new CC algorithm can improve the performance of CC and serves as a competitive solver for large-scale optimization problems. [ABSTRACT FROM AUTHOR]

Two-phase I/O is a well-known strategy for implementing collective MPI-IO functions. It redistributes I/O requests among the calling processes into a form that minimizes the file access costs. As modern parallel computers continue to grow into the exascale era, the communication cost of such request redistribution can quickly overwhelm collective I/O performance. This effect has been observed from parallel jobs that run on multiple compute nodes with a high count of MPI processes on each node. To reduce the communication cost, we present a new design for collective I/O by adding an extra communication layer that performs request aggregation among processes within the same compute nodes. This approach can significantly reduce inter-node communication contention when redistributing the I/O requests. We evaluate the performance and compare it with the original two-phase I/O on Cray XC40 parallel computers (Theta and Cori) with Intel KNL and Haswell processors. Using I/O patterns from two large-scale production applications and an I/O benchmark, we show our proposed method effectively reduces the communication cost and hence maintains the scalability for a large number of processes.

Conversational emotion recognition is a crucial research topic in human-computer interactions. Due to the heavy annotation cost and inevitable label ambiguity, collecting large amounts of labeled data is challenging and expensive, which restricts the performance of current fully-supervised methods in this domain. To address this problem, researchers attempt to distill knowledge from unlabeled data via semi-supervised learning. However, most of these semi-supervised methods ignore multimodal interactive information, although recent works have proven that such interactive information is essential for emotion recognition. To this end, we propose a novel framework to seamlessly integrate semi-supervised learning with multimodal interactions, called “Semi-supervised Multi-modal Interaction Network (SMIN)”. SMIN contains two well-designed semi-supervised modules, “Intra-modal Interactive Module (IIM)” and “Cross-modal Interactive Module (CIM)” to learn intra- and cross-modal interactions. These two modules leverage additional unlabeled data to extract emotion-salient representations. To capture additional contextual information, we utilize the hierarchical recurrent networks followed with the hybrid fusion strategy to integrate multimodal features. These multimodal features are further utilized for conversational emotion recognition. Experimental results on four benchmark datasets (i.e., IEMOCAP, MELD, CMU-MOSI and CMU-MOSEI) demonstrate that SMIN succeeds over existing state-of-the-art strategies on emotion recognition. [ABSTRACT FROM AUTHOR]

Agricultural field polygons within smallholder farming systems are essential to facilitate the collection of geo-spatial data useful for farmers, managers, and policymakers. However, the limited availability of training labels poses a challenge in developing supervised methods to accurately delineate field boundaries using Earth observation (EO) data. This letter introduces an open dataset for training and benchmarking machine learning methods to delineate agricultural field boundaries in polygon format. The large-scale dataset consists of 439 001 field polygons divided into 62 tiles of approximately $5 \times 5$ km distributed across Vietnam and Cambodia, covering a range of fields and diverse landscape types. The field polygons have been meticulously digitized from satellite images, following a rigorous multistep quality control process and topological consistency checks. Multitemporal composites of Sentinel-2 (S2) images are provided to ensure cloud-free data. We conducted an experimental analysis testing a state-of-the-art deep learning (DL) workflow based on fully convolutional networks (FCNs), contour closing, and polygonization. We anticipate that this large-scale dataset will enable researchers to further enhance the delineation of agricultural fields in smallholder farms and to support the achievement of the Sustainable Development Goals (SDGs). The dataset can be downloaded from https://doi.org/10.17026/dans-xy6-ngg6. [ABSTRACT FROM AUTHOR]

In precision agriculture, detecting productive crop fields is an essential practice that allows the farmer to evaluate operating performance separately and compare different seed varieties, pesticides, and fertilizers. However, manually identifying productive fields is often time-consuming, costly, and subjective. Previous studies explore different methods to detect crop fields using advanced machine-learning (ML) algorithms to support the specialists’ decisions, but they often lack good quality labeled data. In this context, we propose a high-quality dataset generated by machine operation combined with Sentinel-2 images tracked over time. As far as we know, it is the first one to overcome the lack of labeled samples by using this technique. In sequence, we apply a semisupervised classification of unlabeled data and state-of-the-art supervised and self-supervised deep-learning (DL) methods to detect productive crop fields automatically. Finally, the results demonstrate high accuracy in positive unlabeled (PU) learning, which perfectly fits the problem where we have high confidence in the positive samples. Best performances have been found in Triplet Loss Siamese given the existence of an accurate dataset and contrastive learning considering situations where we do not have a comprehensive labeled dataset available. [ABSTRACT FROM AUTHOR]

Remote sensing scene classification (RSSC) has recently attracted more attention. However, due to restrictions in the imaging environment and equipment, it is difficult to get a large number of labeled images in remote sensing. This has led to the emergence of few-shot learning for RSSC, which aims to achieve better performance with few labeled samples. Remote sensing images’ large interclass similarity may cause classification confusion. To overcome this issue, this study proposes a double discriminative constraint-based affine nonnegative representation for few-shot RSSC. To be specific, we devise a novel representation-based classifier with two discriminative constraint terms in the objective function and utilize affine nonnegative constraints to restrict the learned parameters. These constraints reduce the correlation between classes and strengthen the class specificity of the learned parameters. Experiments on benchmark datasets demonstrate the effectiveness of our method. [ABSTRACT FROM AUTHOR]

With the high computation efficiency and tracking accuracy, discriminative correlation filters (DCFs) have been applied to unmanned aerial vehicle (UAV) tracking. However, in the scenarios (i.e., complex background and temporary occlusion), DCF-based trackers usually generate low credibility response under the influence of background distractors, which contains multiple side peaks and declines the tracking performance. Motivated by the response consistency in adjacent frames and background information penalization, we propose learning a response interference suppression (RIS) CF to tackle this problem. Specifically, we introduce an RIS regularization into the DCF-based framework, which aims to keep the target area response consistent in adjacent frames and repress distractors’ response in the background. Besides, we adopt a response auxiliary strategy (RAS) to smooth the target response, which intends to obtain the precise location and avoid target drift. Furthermore, extensive experiments on three UAV benchmarks demonstrate the excellent performance of the proposed method against other 19 state-of-the-art trackers. Moreover, the tracking speed of the proposed method can reach $\sim $ 42 frames/s (FPS) on a single CPU. [ABSTRACT FROM AUTHOR]

In this work, Siwa, a micropower 32-bits RISC-V core aimed at implantable medical SoCs is presented. The core was fabricated in a 180-nm CMOS-HV technology to directly drive biological stimuli circuits within the same ASIC. A complete set of power consumption measurements is presented; the core properly operated up to 30 MHz with a current consumption of $52~\mu \text{A}$ /MHz at 1.8-V supply voltage, and < 20-nA leakages at room temperature. Since the existing benchmarks are not completely adequate to compare Siwa performance to other microcontrollers used on implantable medical devices (IMDs), a simple, specific benchmark inspired in a pacemaker operation was developed. The new benchmark considers both the CPU current consumption and performance, sleep and run states, and allows to compare broadly different CPUs and operating conditions for specific IMD applications. Siwa’s performance was compared using this benchmark with 8 and 16-bits MCUs. [ABSTRACT FROM AUTHOR]

Summary: The objective in optimal control problems is to determine the control rule to be applied according to the objective function determined for a given system. Theoretically, different solution methods have been developed starting from calculus of variation such as Pontriagin, Hamilton, or Riccati. In this study, apart from analytical methods, it is proposed to apply algorithms inspired by nature to the same problem. Among these algorithms, genetic algorithm, firefly algorithm, Harris Hawk's optimization algorithm, and differential evolution optimization algorithm have been applied to given optimal control problems and their successes were compared in terms of statistical analysis such as minimum, maximum, average, Wilcoxon, and Friedman analysis. The originality of this study is to control the system with discrete control rule by partitioning the control signal, to be applied to the system which is wanted to be optimally controlled. This will allow the control of different techniques, such as MPC or NMPC, which will contribute to the determination of the global optimum control rule, especially in the solution of nonlinear systems. In addition, three different applications have been performed to demonstrate. As a result of these experiments, it has been shown that these algorithms can be applied successfully to such problems. [ABSTRACT FROM AUTHOR]

Solving constrained multi-objective optimization problems (CMOPs) is one of the most popular research topics in the multi-objective optimization community. Various approaches based on different algorithmic strategies have been proposed for benchmark CMOPs with different features and challenges. Although most of these algorithms employ one or more fixed strategies, determining the most suitable strategy is challenging and problem-dependent. This work presents a general multitasking framework comprising a main task for the original CMOP and an arbitrary number of auxiliary tasks based on different strategies for the auxiliary problems. Reinforcement learning techniques are utilized according to the population state to choose the most suitable auxiliary task during the evolutionary process. As instantiations, Q-Learning and Deep Q-Learning are successfully applied in the framework. Accordingly, two novel algorithms are developed to solve CMOPs. The results demonstrate the effectiveness of the proposed approaches on five benchmark CMOP test suites and real-world problems compared to 11 advanced methods. [ABSTRACT FROM AUTHOR]

Social robots have recently gained popularity, and many human-aware navigation approaches have emerged. This work presents a comprehensive benchmark for quantitatively assessing robot navigation methods. As an automated quantitative approach, our Social Robot Planner Benchmark (SRPB) produces invariable indicators of the algorithm’s performance that can assist the system designer in selecting the best method for the specific application. Our benchmark extends state-of-the-art task performance scores and proposes novel social metrics regarding robot motion naturalness and the perceived safety of humans surrounding the robot. Our social metrics take human tracking reliability into account. Using the SRPB integrated with the TIAGo robot, we assessed the robot’s behaviour operating with traditional and human-aware trajectory planners in simulated and real-world environments. Our experiments tested whether state-of-the-art human-aware trajectory planners significantly improve human-awareness indicators over traditional approaches yet still maintain reasonable navigation performance. An open-source implementation of our benchmark, compatible with the Robot Operating System, is provided.

Phishing attacks have become a perilous threat in recent years, which has led to numerous studies to determine the classification technique that best detects these attacks. Several studies have made comparisons using only specific datasets and techniques without including the most crucial aspect, which is the performance evaluation of data changes. Hence, classification techniques cannot be generalized if they only use specific datasets and techniques. Therefore, this research determined the performance of classification techniques on changing data through a subset of schemes in a dataset. It was conducted using unbalanced and balanced phishing datasets, as well as subset schemes in ratios of 90:10, 80:20, 70:30, and 60:40. The thirteen most recent classification techniques used in preliminary phishing studies were compared and evaluated against ten performance measures. The results showed that the proposed schemes successfully uncover the maximum and minimum performance obtained by a classification technique. These comparisons can provide deeper insights into phishing classification techniques than related research.

New and emerging technologies have recently taken center stage in public policy debates around the world. From the European Union’s (EU) Artificial Intelligence (AI) Act and the Council of Europe’s (CoE) treaty process to soft laws introduced by the Organization for Economic Cooperation and Development (OECD) and a proposal for a bill of rights in the United States , new and emerging technologies have galvanized policymaking at various levels. An overarching objective in such legislative initiatives is upholding human rights not only in the operationalization, but also in the conception and design of new and emerging technologies. But much of the recent legislative and policy initiatives are occurring in the Global North. Partly due to the low uptake of new and emerging technologies, similar initiatives are yet to proliferate in the Global South. In the few jurisdictions where such initiatives exist, one sees clear marks of the regulatory initiatives in the Global North. In Egypt, for instance, a national AI strategy developed based on the international benchmarks mentioned above has recently been tabled. Other countries in Africa have introduced rather generic digital transformation strategies. At the regional level, the African Union appears to pursue two parallel processes aimed at developing a continental AI strategy for Africa in which the influence of AI initiatives from the Global North is inevitable ,. [ABSTRACT FROM AUTHOR]

The electromagnetic levitation (EML) system serves as a key subsystem in maglev trains for the purpose of levitation. It is highly dynamic, open-loop unstable, and safety-critical. The expense of establishing an accurate model out of the Maglev train, in addition to the varying operating conditions, results in an imperfectly known model in engineering practice. Thus high-performance levitation control, w.r.t. an imperfectly known model, is of considerable practical interest. Motivated by such an observation, this article investigates real-time levitation performance optimization of the EML system, with an imperfectly known model. The EML system is first modeled and an equivalent demonstration benchmark is developed. Then, the structure for levitation performance optimization is presented on top of the coprime factorization technique. Furthermore, the real-time levitation performance optimization algorithm is developed, utilizing only the input and output data. In the end, the proposed methods are validated on the benchmark. [ABSTRACT FROM AUTHOR]

We propose a hardware technique for cache coherence over the existing approaches that ensure that shared and less frequently used cache blocks bypass private caches of multiple cores. Furthermore, this manuscript proposes a mechanism to tune the aggressiveness of a data prefetcher. Increased cache hit rate and improved performance have been observed since coherence management and prefetching delays are avoided using the proposed bypassing and thread progress-aware prefetch controlling mechanism. Our approach shows around 19% improvement in cache hit rate and 29% average performance improvement over existing state-of-the-art techniques for Parsec & Splash-2 multithreaded benchmarks. [ABSTRACT FROM AUTHOR]

Several optimization problems have features that hinder the capabilities of searching heuristics. To cope with this issue, different methods have been proposed to manipulate search spaces and improve the optimization process. This paper focuses on Dilation Functions (DFs), which are one of the most promising techniques to manipulate the fitness landscape, by "expanding" or "compressing" specific regions. The definition of appropriate DFs is problem dependent and requires a-priori knowledge of the optimization problem. Therefore, it is essential to introduce an automatic and efficient strategy to identify optimal DFs. With this aim, we propose a novel method based on Genetic Programming, named GP4DFs, which is capable of evolving effective DFs. GP4DFs identifies optimal dilations, where a specific DF is applied to each dimension of the search space. Moreover, thanks to a knowledge-driven initialization strategy, GP4DFs converges to better solutions with a reduced number of fitness evaluations, compared to the state-of-the-art approaches. The performance of GP4DFs is assessed on a set of 43 benchmark functions mimicking several features of real-world optimization problems. The obtained results indicate the suitability of the generated DFs. [ABSTRACT FROM AUTHOR]

Influence maximization is a crucial issue for mining the deep information of social networks, which aims to select a seed set from the network to maximize the number of influenced nodes. To evaluate the influence spread of a seed set efficiently, existing studies have proposed transformations with lower computational costs to replace the expensive Monte Carlo simulation process. These alternate transformations, based on network prior knowledge, induce different search behaviors with similar characteristics to various perspectives. Specifically, it is difficult for users to determine a suitable transformation a priori. This article proposes a multi-transformation evolutionary framework for influence maximization (MTEFIM) with convergence guarantees to exploit the potential similarities and unique advantages of alternate transformations and to avoid users manually determining the most suitable one. In MTEFIM, multiple transformations are optimized simultaneously as multiple tasks. Each transformation is assigned an evolutionary solver. Three major components of MTEFIM are conducted via: 1) estimating the potential relationship across transformations based on the degree of overlap across individuals of different populations, 2) transferring individuals across populations adaptively according to the inter-transformation relationship, and 3) selecting the final output seed set containing all the transformation's knowledge. The effectiveness of MTEFIM is validated on both benchmarks and real-world social networks. The experimental results show that MTEFIM can efficiently utilize the potentially transferable knowledge across multiple transformations to achieve highly competitive performance compared to several popular IM-specific methods. The implementation of MTEFIM can be accessed through the external link in the footnote. [ABSTRACT FROM AUTHOR]

Few-shot object detection (FSOD), which aims at learning a generic detector that can adapt to unseen tasks with scarce training samples, has witnessed consistent improvement recently. However, most existing methods ignore the efficiency issues, e.g., high computational complexity and slow adaptation speed. Notably, efficiency has become an increasingly important evaluation metric for few-shot techniques due to an emerging trend toward embedded AI. To this end, we present an efficient pretrain-transfer framework (PTF) baseline with no computational increment, which achieves comparable results with previous state-of-the-art (SOTA) methods. Upon this baseline, we devise an initializer named knowledge inheritance (KI) to reliably initialize the novel weights for the box classifier, which effectively facilitates the knowledge transfer process and boosts the adaptation speed. Within the KI initializer, we propose an adaptive length re-scaling (ALR) strategy to alleviate the vector length inconsistency between the predicted novel weights and the pretrained base weights. Finally, our approach not only achieves the SOTA results across three public benchmarks, i.e., PASCAL VOC, COCO and LVIS, but also exhibits high efficiency with $1.8-100\times $ faster adaptation speed against the other methods on COCO/LVIS benchmark during few-shot transfer. To our best knowledge, this is the first work to consider the efficiency problem in FSOD. We hope to motivate a trend toward powerful yet efficient few-shot technique development. The codes are publicly available at https://github.com/Ze-Yang/Efficient-FSOD. [ABSTRACT FROM AUTHOR]

Feature-based domain adaptation methods project samples from different domains into the same feature space and try to align the distribution of two domains to learn an effective transferable model. The vital problem is how to find a proper way to reduce the domain shift and improve the discriminability of features. To address the above issues, we propose a unified Probability-based Graph embedding Cross-domain and class Discriminative feature learning framework for unsupervised domain adaptation (PGCD). Specifically, we propose novel graph embedding structures to be the class discriminative transfer feature learning item and cross-domain alignment item, which can make the same-category samples compact in each domain, and fully align the local and global geometric structure across domains. Besides, two theoretical analyses are given to prove the interpretability of the proposed graph structures, which can further describe the relationships between samples to samples in single-domain and cross-domain transfer feature learning scenarios. Moreover, we adopt novel weight strategies via probability information to generate robust centroids in each proposed item to enhance the accuracy of transfer feature learning and reduce the error accumulation. Compared with the advanced approaches by comprehensive experiments, the promising performance on the benchmark datasets verify the effectiveness of the proposed model. [ABSTRACT FROM AUTHOR]

This article presents a static deadlock analysis for single-path MPI programs. Deadlock is when processes are blocked indefinitely by a circular communication dependency. A single path program is one that does not decode messages for control flow. The analysis records a program execution in the form of a trace and then determines from that trace whether there exists any feasible deadlocking schedules. The primary contribution is the combining of identical consecutive sends or receives into single macro actions. This simplified trace is analyzed for potential deadlock cycles. An abstract machine identifies infeasible cycles, and those not identified by the machine are encoded as satisfiability problems for an SMT solver to resolve. The action combination reduces the complexity of identifying and filtering cycles before needing the costly SMT solver. This article shows the effectiveness of the action combination in experiments on a benchmark suite comparing to traces without action combination and other state-of-the-art deadlock analyses. [ABSTRACT FROM AUTHOR]

In the conventional studies related to steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs), the window length (detection time) was typically predetermined through the offline analysis, which had limitations of practical applicability of a BCI system due to the inter-subject/trial variability of electroencephalography (EEG) signals. To address these limitations, this study aims to automatically optimize the window length for each trial based on training-free approaches and proposes a novel adaptive window method (ANCOVA-based filter-bank canonical correlation analysis, ABFCCA) for SSVEP-based BCIs. The proposed method is based on analysis of covariance (ANCOVA) which is applied after feature extraction by the conventional training-free SSVEP recognition approaches. To evaluate the performance of the proposed method, conventional fixed window and recent adaptive window methods were compared using two open-access datasets. In the Benchmark dataset, the average information transfer rate (ITR) was 146.81 bits/min, the average accuracy 93.55%, and the average window length 1.53 s. In the OpenBMI dataset, the average ITR was 119.01 bits/min, the average accuracy 83.50%, and the average window length 0.65 s. The proposed method significantly outperformed the conventional approaches with fixed window in terms of the accuracy and ITR, and is applicable to various SSVEP-based BCI paradigms based on the criterion of significance level without offline analysis to find optimal hyper-parameters. ABFCCA is enabled the practical use of various BCI systems by automatically optimizing the window length independently. [ABSTRACT FROM AUTHOR]

Zero-Shot Learning (ZSL) aims to recognize unseen classes that never appear during training. Recently, generative adversarial networks (GANs) have been introduced to convert ZSL into a supervised learning problem by synthesizing unseen visual features. However, since unseen classes are never experienced for the generator during training, the synthesized unseen visual features often become heavily biased towards seen classes, or sometimes there is even no meaningful class that can be assigned to them. This is known as the bias problem. In this paper, we propose a novel method, namely Adaptive Bias-Aware GAN (ABA-GAN), to alleviate generating biased visual features. For this purpose, we build a semantic adversarial network to regularize the feature generator. Specifically, an adaptive adversarial loss is proposed to constrain the feature distributions, which avoids the generation of meaningless visual features. Meanwhile, a domain divider is presented to explicitly distinguish synthesized visual features between seen and unseen domains, such that the bias towards seen classes can be alleviated. Moreover, we propose a novel metric named bias score (BS) to explicitly quantify the degree of the strong bias. Extensive experiments on four widely used benchmark datasets demonstrate that our proposed method outperforms the state-of-the-art approaches under both ZSL and GZSL protocols. [ABSTRACT FROM AUTHOR]

Weakly supervised object detection (WSOD) aims to train object detectors by using only image-level annotations. Many recent works on WSOD adopt multiple instance detection networks (MIDN), which usually generate a certain number of proposals and regard proposal classification as a latent model learning within image classification. However, these methods tend to detect salient object, salient object parts and clustered objects due to lack of instance-level annotations during training. Thus a core issue is how to guarantee that the network learn as many objects with precise bounding boxes as possible. In this paper, we address this issue by exploiting the potential of proposal scores during training. We propose an adaptive instance refinement (AIR) framework with three novel designs, which can be integrated with MIDN into a single network. Specifically, adaptive instance mining attempts to discover all positive instances according to the score distribution of proposals and their spatial similarity. Adaptive score modulation dynamically adjusts proposal scores to make the network focus more on instances with different difficulties in different training iterations. Adaptive knowledge refinement distills important information from all previous stages by the weighted average of proposal scores. The experimental results on the PASCAL VOC 2007 and 2012 benchmarks and the MS COCO benchmark demonstrate that AIR significantly improves the performance of the original MIDN and achieves the state-of-the-art results. [ABSTRACT FROM AUTHOR]

Since light is scattered and absorbed by water, underwater images have inherent degradation (e.g., hazing, color shift), consequently impeding the development of remotely operated vehicles (ROVs). Toward this end, we propose a novel method, referred to as ${B}$ est ${o}\text{f}$ Bo th World ${s}$ (Boths). With parameters of only 0.0064 M, Boths can be considered a super lightweight neural network for underwater image enhancement. On the whole, it has three levels: structure and detail features; pixel and channel dimensions; high- and low-frequency information. Each of these three levels represents “Best of Both Worlds.” Initially, by interacting with structure and detail features, Boths can focus on these two aspects at the same time. Further, our network can simultaneously consider channel and pixel dimensions through 3-D attention learning, which is more similar to human visual perception. Lastly, the proposed model can focus on high- and low-frequency information, through a novel loss function based on the wavelet transforms. Upon subsequent analysis and evaluation, Boths has shown superior performance compared with state-of-the-art (SOTA) methods. Our models and datasets are publicly available at: https://github.com/perseveranceLX/Boths. [ABSTRACT FROM AUTHOR]