Code Vectors Research Articles

Dxtb-An efficient and fully differentiable framework for extended tight-binding.

Automatic differentiation (AD) emerged as an integral part of machine learning, accelerating model development by enabling gradient-based optimization without explicit analytical derivatives. Recently, the benefits of AD and computing arbitrary-order derivatives with respect to any variable were also recognized in the field of quantum chemistry. In this work, we present dxtb-an open-source, fully differentiable framework for semiempirical extended tight-binding (xTB) methods. Developed entirely in Python and leveraging PyTorch for array operations, dxtb facilitates extensibility and rapid prototyping while maintaining computational efficiency. Through comprehensive code vectorization and optimization, we essentially reach the speed of compiled xTB programs for high-throughput calculations of small molecules. The excellent performance also scales to large systems, and batch operability yields additional benefits for execution on parallel hardware. In particular, energy evaluations are on par with existing programs, whereas the speed of automatically differentiated nuclear derivatives is only 2 to 5 times slower compared to their analytical counterparts. We showcase the utility of AD in dxtb by calculating various molecular and spectroscopic properties, highlighting its capacity to enhance and simplify such evaluations. Furthermore, the framework streamlines optimization tasks and offers seamless integration of semiempirical quantum chemistry in machine learning, paving the way for physics-inspired end-to-end differentiable models. Ultimately, dxtb aims to further advance the capabilities of semiempirical methods, providing an extensible foundation for future developments and hybrid machine learning applications. The framework is accessible at https://github.com/grimme-lab/dxtb.

Automated patch correctness predicting to fix software defect

Automated Program Repair (APR) is a technique that can automatically fix software defects without manual debugging, playing a crucial role in software development and maintenance. However, the patches generated by APR still suffer from the problem of overfitting, which poses a significant threat to practical applications. Previous studies have proposed various approaches to predict the correctness of the patch to address this issue, primarily including static-based methods, dynamic-based methods, and learning-based methods. However, these methods all have their own limitations, making it difficult to accurately extract code semantic features and achieve comprehensive prediction. To address the aforementioned challenges, we propose a learning-based unsupervised classification model, Automated Patch cOrrectness aSsessmenT based on muLtiple pErspectives (APOSTLE), for predicting the correctness of patches. Specifically, APOSTLE consists of three components: code vectorization component, where advanced pre-trained models are used to achieve efficient extraction of semantic features from the code; similarity and code change degree calculation component, where APOSTLE calculates the similarity and the degree of change to the code; comprehensive evaluation component, where APOSTLE conducts comprehensive evaluation, addressing the issue of prediction comprehensiveness. Experiments on a collection of 1278 patches (written by developers or generated by 32 APR tools) demonstrate that APOSTLE achieves an AUC value of 0.801, an MAP value of 0.855, and an MRR value of 0.944, outperforming the state-of-the-art approach BATS by 8.3%, 6.0%, and 9.0%, respectively. APOSTLE successfully achieves accurate extraction of code semantic features while achieving comprehensive patch correctness prediction.

Automatic purification of skylight spectrum based on an autoencoder

Abstract In the realm of astronomical spectroscopic observation, the purity of skylight spectra is crucial for accurate analysis, often complicated by interference from neighboring celestial objects. Addressing this challenge, a novel automatic purification algorithm is proposed in this study, leveraging the power of auto-coding mechanisms. By harnessing the inherent data features and shared representations between skylight and target objects, this method effectively isolates skylight spectra from residual signals of neighboring objects. Central to this approach is the utilization of an encoder-decoder framework. The encoder dynamically learns the combined features of skylight and target objects, generating respective coding vectors. These vectors are then utilized to subtract the target-specific features from the skylight coding space, facilitating the extraction of pure skylight characteristics. Subsequently, the decoder reconstructs the refined skylight data, effectively eliminating residual components associated with neighboring celestial bodies. A notable strength of this algorithm lies in its ability to perform feature extraction and purification autonomously, without relying on labeled data. Experimental validation conducted on LAMOST-DR5 datasets demonstrates the efficacy of the proposed method. By effectively removing spectral information from target objects within skylight spectra, the algorithm yields comparatively pure skylight spectra, paving the way for enhanced astronomical spectroscopic analyses.

Boosting Metamorphic Relation Prediction via Code Representation Learning: An Empirical Study

AbstractMetamorphic testing (MT) is an effective testing technique having a broad range of applications. One key task for MT is the identification of metamorphic relations (MRs), which is a fundamental mechanism in MT and is critical to the automation of MT. Prior studies have proposed approaches for predicting MRs (PMR). One major idea behind these PMR approaches is to represent program source code information via manually designed code features and then to apply machine‐learning–based classifiers to automatically predict whether a specific MR can be applied on the target program. Nevertheless, the human‐involved procedure of selecting and extracting code features is costly, and it may not be easy to obtain sufficiently comprehensive features for representing source code. To overcome this limitation, in this study, we explore and evaluate the effectiveness of code representation learning techniques for PMR. By applying neural code representation models for automatically mapping program source code to code vectors, the PMR procedure can be boosted with learned code representations. We develop 32 PMR instances by, respectively, combining 8 code representation models with 4 typical classification models and conduct an extensive empirical study to investigate the effectiveness of code representation learning techniques in the context of MR prediction. Our findings reveal that code representation learning can positively contribute to the prediction of MRs and provide insights into the practical usage of code representation models in the context of MR prediction. Our findings could help researchers and practitioners to gain a deeper understanding of the strength of code representation learning for PMR and, hence, pave the way for future research in deriving or extracting MRs from program source code.

Quantum Reduction of Finding Short Code Vectors to the Decoding Problem

Quantum Reduction of Finding Short Code Vectors to the Decoding Problem

Abnormal Traffic Detection System Based on Feature Fusion and Sparse Transformer

This paper presents a feature fusion and sparse transformer-based anomalous traffic detection system (FSTDS). FSTDS utilizes a feature fusion network to encode the traffic data sequences and extracting features, fusing them into coding vectors through shallow and deep convolutional networks, followed by deep coding using a sparse transformer to capture the complex relationships between network flows; finally, a multilayer perceptron is used to classify the traffic and achieve anomaly traffic detection. The feature fusion network of FSTDS improves feature extraction from small sample data, the deep encoder enhances the understanding of complex traffic patterns, and the sparse transformer reduces the computational and storage overhead and improves the scalability of the model. Experiments demonstrate that the number of FSTDS parameters is reduced by up to nearly half compared to the baseline, and the success rate of anomalous flow detection is close to 100%.

160 Gbps MMF/FSO system based on OAM beams and PV code under rainy weather

In this paper, a high-speed transmission system that combines a multimode fiber (MMF) with Free Space Optics (FSO) is proposed to accommodate the rapid growth of traffic. As Orbital Angular Beams Multiplexing (OAM) plays a crucial role in 6G networks by enhancing transmission capacity, four OAM beams (LG0,0\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${LG}_{\\mathrm{0,0}}$$\\end{document}, LG0,20\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${LG}_{\\mathrm{0,20}}$$\\end{document}, LG0,40\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${LG}_{\\mathrm{0,40}}$$\\end{document}, and LG0,40\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${LG}_{\\mathrm{0,40}}$$\\end{document}) are utilized in this study. Additionally, the Optical Code Division Multiple Access (OCDMA) technique is employed, known for its high level of confidentiality and the ability to allow multiple channels to transmit data simultaneously. To ensure ubiquitous data transmission, two channels: MMF and FSO are used. Accordingly, in our proposed MMF/FSO system, we employ four OAM beams, with each carrying four OCDMA channels assigned distinct Permutation Vector (PV) codes. With a data transmission rate of 10 Gbps on each OCDMA channel, the overall capacity of the system amounts to 160 Gbps (10 Gbps × 4 OCDMA channels × 4 OAM beams). Moreover, the performance is investigated and evaluated using two scenarios. The first assumes an FSO range of 5 km and Clear Air (CA) weather conditions, with a variable MMF length. In contrast, the second scenario maintains a fixed MMF length of 1 km while altering the FSO span. Weather conditions like CA and rain conditions; Light Rain (LR), Medium Rain (MR), and Heavy Rain (HR) are considered while evaluating the system performance in the second case. The bit error rate (BER), travelling distances in the channel either MMF or FSO, and eye diagrams are among the metrics used for evaluation. They all provide information about the received signal performance. Finally, the obtained results for the proposed MMF/FSO system, utilizing OAM beams and PV codes, indicate that in the first scenario, the system can support a total capacity of 160 Gbps across all channels. The performance is noteworthy, with a BER below 10-5\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${10}^{-5}$$\\end{document} and a wide eye opening, enabling successful transmission over 6.2 km (1.2 km MMF length + fixed 5 km FSO span). For the second scenario, the achievable distance is decreased due to the attenuation caused by rain. It becomes 2.62 km (1 km MMF + 1.62 km FSO span), 5.2 km (1 km MMF + 1.2 km FSO span), and 1.77 km (1 km MMF + 0.77 km FSO span), under LR, MR, and HR, respectively.

Research of cognitive data processing in radio communication systems with permutation decoding

Background. The need to use permutation decoding tools in radio communication systems is explained by the increased error correction capabilities of this method. In this case, complex matrix calculations during the search for equivalent codes according to the classical scheme of permutation decoding are replaced by a list of ready-made solutions. These solutions are calculated a priori and entered into the cognitive cards of the decoder processor, which makes the method a convenient tool in the procedure for ensuring information reliability when controlling, for example, unmanned vehicles via radio channels. In fact, matrix calculations on board are replaced by searching the list of cognitive maps for the right solution corresponding in real time to the current permutation of reliable character numerators. However, data processing in the decoder’s cognitive map requires a special description.
 Aim. The study of methods for identifying permutations of character numerators of code vectors in order to effectively transform them in a system of cognitive maps of a permutation decoder.
 Methods. The paper reveals the subtle structure of cognitive maps of productive and unproductive permutations of numerators, which allows on a regular basis to obtain an alternative solution for switching to a set of productive permutations when the receiver receives an unproductive permutation, thereby excluding the use of trial and error.
 Results. The efficiency of the permutation decoder increases due to the implementation of permutations that were originally included in a set of solutions introduced into the cognitive map of unproductive permutations.
 Conclusion. A family of microcontrollers is proposed to implement the principle of interaction of cognitive maps with a system of alternative solutions.

Deep code search efficiency based on clustering

AbstractThe deep‐learning based code search model mainly takes accuracy as the only target for judging the performance of the model, ignoring the efficiency of code search. This article proposes a clustering‐based code search model (C‐DCS). C‐DCS uses the K‐Means to divide the code vector base into K clusters and obtains the center vectors of K clusters. While searching, C‐DCS first matches the query vector with the K center vectors to get the best matching center vector. After matching the center vector, C‐DCS matches the query vector with code vectors in the cluster corresponding to the best matching center vector one by one and then gets the best matching code snippet vector. To verify the efficiency of C‐DCS in the code search task, experimental analysis was built on a large dataset. The experimental results showed that C‐DCS saves 92.2% of the search time compared to the baseline model while remaining the accuracy. In the experimental evaluation section, we optimized the K‐Means algorithm to improve the code search efficiency of C‐DCS further, reducing the search time to 93.8% of the baseline model. Hence, C‐DCS reduces the code search time greatly with not affecting the accuracy, improving the efficiency of software development.

Novel compressed linear network coding vectors for multihop communication networks

Novel compressed linear network coding vectors for multihop communication networks

Autovesk: Automatic Vectorized Code Generation from Unstructured Static Kernels Using Graph Transformations

Leveraging the SIMD capability of modern CPU architectures is mandatory to take full advantage of their increased performance. To exploit this capability, binary executables must be vectorized, either manually by developers or automatically by a tool. For this reason, the compilation research community has developed several strategies for transforming scalar code into a vectorized implementation. However, most existing automatic vectorization techniques in modern compilers are designed for regular codes, leaving irregular applications with non-contiguous data access patterns at a disadvantage. In this article, we present a new tool, Autovesk, that automatically generates vectorized code from scalar code, specifically targeting irregular data access patterns. We describe how our method transforms a graph of scalar instructions into a vectorized one, using different heuristics to reduce the number or cost of instructions. Finally, we demonstrate the effectiveness of our approach on various computational kernels using Intel AVX-512 and ARM SVE. We compare the speedups of Autovesk vectorized code over GCC, Clang LLVM, and Intel automatic vectorization optimizations. We achieve competitive results on linear kernels and up to 11× speedups on irregular kernels.

Methodological approach for the automatic discrimination of pictorial materials using fused hyperspectral imaging data from the visible to mid-infrared range coupled with machine learning methods

Hyperspectral imaging represents a powerful tool for the study of artwork’s materials since it permits to obtain simultaneously information about the spectral behavior of the materials and their spatial distribution. By combining hyperspectral images performed on several spectral intervals (visible, near infrared and mid-infrared ranges) through chemometric methods it is possible to clearly identify most of the materials used in painting (i.e., pigments, dyes, varnishes, and binders). Moreover, in the last decade, the development of machine learning algorithms coupled with comprehensive and continuously updated databases opens new perspective on the automatic recognition of pictorial materials.In this work, we propose a novel procedure to support the automatic discrimination of pictorial materials consisting in a mid-level data fusion on imaging datasets coming from two commercial hyperspectral cameras, in the 400–1000 nm and 1000–2500 nm spectral ranges, respectively, and a MAcroscopic Fourier Transform InfRared scanning in reflection mode (MA-rFTIR), in the 7000 to 350 cm-1 (1428 nm − 28 μm) spectral range. The automatic recognition of 102 pictorial mock-ups from the fused data is performed by testing the performance of ECOC-SVM (error-correcting output coding and support vector machine) model obtaining a good predictive result with only few pixels that are confused with other classes. The methodology described in this paper demonstrates that an accurate paint layer multiclass recognition is feasible, and the use of chemometric approaches solves some challenges involving the study of materials.

Распределение мощности кодов с наименьшей избыточностью алфавитов в зависимости от количества бит и кодового расстояния

Purpose: To investigate the dependence of the maximum power of codes on the number of digits and the minimum code distance; to find an approach to determine the optimal rules for constructing the check vector of a separable code from the point of view of ensuring minimal redundancy with a given reliability of message transmission. Methods: Computer simulation has been used to conduct experimental studies. For theoretical studies, the method of analytical review, graph theory, and coding theory have been applied. Results: Theoretical and experimental studies have obtained certain specific cases of power distribution for code alphabets with a given Hamming distance for various constant lengths, generated using the previously described algorithm. A method for doubling the power of arbitrary binary codes is proposed and described, as well as a method for obtaining codes with the least redundancy of powers M=2f, where f is a natural number for a predetermined minimum code distance by recursively using the method proposed in the article. Practical significance: An algorithm has been developed for doubling the power of the code alphabet while maintaining the required reliability of data transmission. A technique for analyzing the resulting matrices of code vectors is obtained in order to determine the rules for calculating check bits without using cyclic algorithms.

Hybrid character sums and near-optimal partial Hadamard codebooks

In this paper, we first study exponential sums of the form∑x∈Fqψ(f(x))χ(g(x)) called hybrid character sums, where f,g are functions over finite field Fq, ψ is a nontrivial multiplicative character and χ is a nontrivial additive character of Fq. The absolute values or explicit values of them are determined under certain conditions. We then present several constructions of complex codebooks which are also called frames. The proposed codebooks are constructed from Hadamard matrices with binary row selection sequences. The hybrid character sums play an important role in determining the maximum magnitude of the inner products between vectors in these codebooks. It is proved that the magnitude of the inner products between distinct code vectors of each codebook is two-valued, and the maximum magnitude of each codebook asymptotically achieves the Welch bound equality. The constructed codebooks are equivalent to nearly equiangular tight frames and have nice application in compressed sensing matrices.

Improvements to code2vec: Generating path vectors using RNN

Source code analysis has many application scenarios, such as code plagiarism detection and software vulnerability search. Source code analysis can benefit from machine learning, but it typically requires a standard vector representation and cannot be directly applied to the source code. Thus, we are required to embed source code into vector representation while maintaining the semantics of the code as much as possible. Code2vec proposes a code embedding method that converts source code into code vector through Abstract Syntax Tree(AST). However, we found that code2vec uses a hashing algorithm to generate the identifier for the path in the path context, which leads to the loss of node information in the path and also causes the model training parameters to be very large. Therefore, we present a new path representation which utilizes RNN to generate vectors for paths. We also proposed alternative model designs and evaluated their impact on the model in the experiments. The results we obtained in a challenging source code classification task suggest that, compared to code2vec, the RNN-based paths representation can produce a better embedding model with fewer training parameters.

The plasma–sheath transition and Bohm criterion in a high recycling divertor

The high recycling regime of a divertor is characterized by high plasma particle fluxes and low temperature at the target, where a strong hydrogen recirculation loop exists. Atomic processes in the high recycling regime, such as ion–neutral friction and radiation, can affect the plasma momentum and energy transport in the sheath transition region. Here, the plasma–sheath transition near a high recycling wall is investigated. The Bohm speed, which constraints the ion exit flow speed, is evaluated from a transport model that accounts for the effect of the anisotropic transport and atomic collisions in the transition layer. A first principles kinetic code vector particle-in-cell with the atomic collision package is used to investigate a 1D self-consistent slab plasma with a high recycling boundary for the tungsten and carbon divertors. The results demonstrate the accuracy of the Bohm speed model in predicting the ion exit flow speed in the transition region, as well as the reduction of the Bohm speed due to the ion–neutral friction. The effect of different wall materials, tungsten, and carbon, on the Bohm speed and near-wall plasma profile is shown.

Impact of COVID-19 Vaccination on Pregnant Women.

In light of the COVID-19 pandemic, researchers across the world hastened to develop vaccines that would aid in bolstering herd immunity. Utilizing mRNA coding and viral vector technology, the currently approved vaccines were required to undergo extensive testing to confirm their safety for mass usage in the general population. However, clinical trials failed to test the safety and efficacy of the COVID-19 vaccines in groups with weakened immune systems, especially pregnant women. Lack of information on the effects of vaccinations in pregnancy and the safety of fetuses are among the topmost reasons preventing pregnant women from receiving immunization. Thus, the lack of data examining the effects of COVID-19 vaccinations on pregnant women must be addressed. This review focused on the safety and efficacy of the approved COVID-19 vaccinations in pregnancy and their impact on both maternal and fetal immune responses. For that, we took the approach of combined systematic review/meta-analysis and compiled the available data from the original literature from PubMed, Web of Science, EMBASE and Medline databases. All articles analyzed presented no adverse effects of vaccination in pregnancy, with varying conclusions on the degree of effectiveness. The majority of the findings described robust immune responses in vaccinated pregnant women, successful transplacental antibody transfer, and implications for neonatal immunity. Hence, findings from the cumulative data available can be helpful in achieving COVID-19 herd immunization, including pregnant women.

Bird sound classification based on ECOC-SVM

Birds are an important part of the ecosystem. Knowing the number of existing bird species is important for maintaining ecological balance. The sounds of birds contain rich biological features. The classification of bird species based on their sounds helps to protect the ecological environment. In this study, a fused classification method based on the Error Correction Output Coding (ECOC) and Support Vector Machines (SVM) is investigated to classify 11 species of birds. The Mel Frequency Cepstrum Coefficients (MFCC) of bird sound are extracted as acoustic feature. The ECOC-SVM is compared with Random Forest (RF), Gaussian Mixture Model (GMM) and a multi-layer perceptron neural network (MLP-NN) and a convolutional neural network (CNN) for the bird sound classification. The proposed ECOC-SVM model is validated on the open dataset available from the Macaulay Library at Cornell University. The ECOC-SVM achieves the best classification result and relatively low computational cost among all discussed classifiers on the test set with 100 % accuracy and 52 ms prediction time. The results show that the proposed ECOC-SVM model could well apply on the bird sound classification of multiple species with excellent performance and relatively low computational cost.

6G enabling FSO communication system employing integrated PDM-OAM-OCDMA transmission: impact of weather conditions inIndia.

Currently, free space optics (FSO) transmission has gained attention due to its capability to deliver high data rates with a high level of security. In addition, using multiplexing techniques with FSO makes it capable of handling the excessive increase in traffic data and supporting the 5G and 6G wireless network requirements. Accordingly, in this paper, a novel, to the best of our knowledge, FSO system integrating three multiplexing techniques-orbital angular momentum (OAM) (four OAM beams are used) polarization-division multiplexing (two polarization states are used), and optical code division multiple access (four channels assigned with permutation vector codes are used)-is proposed. Additionally, the effects of rainy, foggy, dusty weather and the real weather for four different Indian cities that have different geographical locations are studied and investigated. The system performance is evaluated based on the bit error rate (BER), quality factor (Q-factor), maximum FSO range, and eye diagram opening. The simulation results show successful transmission of 320Gbps overall capacity with a maximum achievable FSO range of 7km under clear weather. On the other hand, the shortest FSO range of 0.105km is achieved when there are heavy dust storms. As for the Indian cities, Srinagar (hilly area of India) achieves the shortest FSO link, which is 4.2km while the largest range of 7km is observed for Chennai city (coastal area of India). All these ranges are evaluated for a log(BER) value <-7. Consequently, this proposed transmission model is suggested for use in 6G applications of FSO communication systems.

Excalibur

In recent years, hardware has become increasingly diverse, in terms of features as well as performance. This poses a problem for complex software in general and database systems in particular. To achieve top-notch performance, we need to exploit hardware features, but do not fully know which behave best on the current, and more-so future, machines. Specializing query execution methods for many diverse hardware platforms will significantly increase database software complexity and also poses a physical query optimization problem that cannot be solved robustly with static cost models. In this paper, we propose a new query execution architecture addressing these problems. Based on the flexible domain-specific language VOILA, it can generate thousands of different flavors from a single code-base. As an abstraction, a virtual machine (VM) allows hiding physical execution details, such that the VM can transparently switch between different execution tactics within each query, applied at a fine granularity. We show rules to describe a search space for good tactics, and describe efficient search strategies, that limit the overhead of adaptive JIT code generation and compilation. The VM starts executing each query in full vectorized code style, but adaptively replaces (parts of) query pipelines by code fragments compiled using different execution flavors, exploring this search space and exploiting the best tactics found, casting adaptive query execution into a Multi-Armed Bandit (MAB) problem. Excalibur, our prototype, outperforms open-source systems by up to 28X and the state-of-the-art system Umbra by up to 1.8X. In specific queries Excalibur performs up to 2X faster than static flavors.