Performance Improvement Research Articles

Causal Representation Learning via Counterfactual Intervention

Existing causal representation learning methods are based on the causal graph they build. However, due to the omission of bias within the causal graph, they essentially encourage models to learn biased causal effects in latent space. In this paper, we propose a novel causally disentangling framework that aims to learn unbiased causal effects. We first introduce inductive and dataset biases into traditional causal graph for the physical concepts of interest. Then, we eliminate the negative effects from these two biases by counterfactual intervention with reweighted loss function for learning unbiased causal effects. Finally, we employ the causal effects into the VAE to endow the latent representations with causality. In particular, we highlight that removing biases in this paper is regarded as a part of learning process for unbiased causal effects, which is crucial for causal disentanglement performance improvement. Through extensive experiments on real-world and synthetic datasets, we show that our method outperforms different baselines and obtains the state-of-the-art results for achieving causal representation learning.

Continuous Rotation Group Equivariant Network Inspired by Neural Population Coding

Neural population coding can represent continuous information by neurons with a series of discrete preferred stimuli, and we find that the bell-shaped tuning curve plays an important role in this mechanism. Inspired by this, we incorporate a bell-shaped tuning curve into the discrete group convolution to achieve continuous group equivariance. Simply, we modulate group convolution kernels by Gauss functions to obtain bell-shaped tuning curves. Benefiting from the modulation, kernels also gain smooth gradients on geometric dimensions (e.g., location dimension and orientation dimension). It allows us to generate group convolution kernels from sparse weights with learnable geometric parameters, which can achieve both competitive performances and parameter efficiencies. Furthermore, we quantitatively prove that discrete group convolutions with proper tuning curves (bigger than 1x sampling step) can achieve continuous equivariance. Experimental results show that 1) our approach achieves very competitive performances on MNIST-rot with at least 75% fewer parameters compared with previous SOTA methods, which is efficient in parameter; 2) Especially with small sample sizes, our approach exhibits more pronounced performance improvements (up to 24%); 3) It also has excellent rotation generalization ability on various datasets such as MNIST, CIFAR, and ImageNet with both plain and ResNet architectures.

A Primal-Dual Algorithm for Hybrid Federated Learning

Very few methods for hybrid federated learning, where clients only hold subsets of both features and samples, exist. Yet, this scenario is very important in practical settings. We provide a fast, robust algorithm for hybrid federated learning that hinges on Fenchel Duality. We prove the convergence of the algorithm to the same solution as if the model was trained centrally in a variety of practical regimes. Furthermore, we provide experimental results that demonstrate the performance improvements of the algorithm over a commonly used method in federated learning, FedAvg, and an existing hybrid FL algorithm, HyFEM. We also provide privacy considerations and necessary steps to protect client data.

Parallel Beam Search Algorithms for Domain-Independent Dynamic Programming

Domain-independent dynamic programming (DIDP), a model-based paradigm based on dynamic programming, has shown promising performance on multiple combinatorial optimization problems compared with mixed integer programming (MIP) and constraint programming (CP). The current DIDP solvers are based on heuristic search, and the state-of-the-art solver, complete anytime beam search (CABS), uses beam search. However, the current DIDP solvers cannot utilize multiple threads, unlike state-of-the-art MIP and CP solvers. In this paper, we propose three parallel beam search algorithms and develop multi-thread implementations of CABS. With 32 threads, our multi-thread DIDP solvers achieve 9 to 39 times speedup on average and significant performance improvement over the sequential solver, finding the new best solutions for two instances of the traveling salesperson problem with time windows. In addition, our solvers outperform multi-thread MIP and CP solvers in four of the six combinatorial optimization problems evaluated.

An Instruction Inflation Analyzing Framework for Dynamic Binary Translators

Dynamic binary translators (DBTs) are widely used to migrate applications between different instruction set architectures (ISAs). Despite extensive research to improve DBT performance, noticeable overhead remains, preventing near-native performance, especially when translating from complex instruction set computer (CISC) to reduced instruction set computer (RISC). For computational workloads, the main overhead stems from translated code quality. Experimental data show that state-of-the-art DBT products have dynamic code inflation of at least 1.46. This indicates that on average, more than 1.46 host instructions are needed to emulate one guest instruction. Worse, inflation closely correlates with translated code quality. However, the detailed sources of instruction inflation remain unclear. To understand the sources of inflation, we present Deflater , an instruction inflation analysis framework comprising a mathematical model, a collection of black-box unit tests called BenchMIAOes , and a trace-based simulator called InflatSim . The mathematical model calculates overall inflation based on the inflation of individual instructions and translation block optimizations. BenchMIAOes extract model parameters from DBTs without accessing DBT source code. InflatSim implements the model and uses the extracted parameters from BenchMIAOes to simulate a given DBT’s behavior. Deflater is a valuable tool to guide DBT analysis and improvement. Using Deflater, we simulated inflation for three state-of-the-art CISC-to-RISC DBTs: ExaGear, Rosetta2, and LATX, with inflation errors of 5.63%, 5.15%, and 3.44%, respectively for SPEC CPU 2017, gaining insights into these commercial DBTs. Deflater also efficiently models inflation for the open source DBT QEMU and suggests optimizations that can substantially reduce inflation. Implementing the suggested optimizations confirms Deflater’s effective guidance, with 4.65% inflation error, and gains 5.47x performance improvement.

Performance improvement of radial distribution system with distributed generation and intermittent load

The distribution network's load situation is typically variable and non-uniform. Therefore, a modified Distributed Generation (DG) algorithm is proposed in this paper, which uses the analytical approach and applies to all kinds of radial distribution feeder networks. The specified parameter of the distribution network is given accurately to the programme manually or through the file uploading mode, and the power flow analysis is done by MATLAB software. The results are obtained, including the specific location of DG placement which is optimised analytically. Here, the voltage and power profiles are monitored seriously, and the minimisation of voltage drop, and power loss is obtained, which is precise of a high value. The system is implemented for the IEEE 33 bus system which is got from Panchasakha Feeder Network. In this proposed method, DGs are deployed to reduce the loss in the transmission line as well as to maintain the voltage stability in the system by an analytical algorithm approach.

Thermal performance and experimental analysis of stainless steel flat plate solar collector with full-flow channels

The thermal performance of a flat plate solar collector (FPSC) is a critical indicator that depends on the environment, operational parameters, and dimensions. This study examines the impact of size on thermal performance improvement mechanisms. Firstly, numerical simulation models are introduced as the foundation for optimization research. This involves analyzing the flow resistance of microchannels and defining their structural parameters. Furthermore, experimental tests were conducted on a stainless steel flat plate solar collector (S/S FPSC) with the best design parameters to validate the accuracy of the mathematical model during the design phase. The results indicate that increasing the width of the microchannel and the height of corrugations can effectively enhance the thermal performance of the S/S FPSC. The momentary efficiency is projected to reach a remarkable 86.10% under ideal circumstances. Additionally, a mathematical expression was proposed to establish the relationship between the surrounding conditions and the momentary efficiency of the S/S FPSC. Moreover, the microchannel comprises S/S material, maintaining a homogeneous temperature distribution to maximize heat absorption. The use of stainless steel also extends the lifespan of the FPSC.

A combination prediction model based on Theil coefficient and induced continuous aggregation operator for the prediction of Shanghai composite index

This paper proposes an interval combination prediction model for Shanghai composite index, utilizing the Theil coefficient and the induced continuous generalized ordered weighted logarithmic harmonic averaging (ICGOWLHA) operator. The effectiveness of the proposed model under specific weight conditions and the existence of its analytical solution are demonstrated. Shanghai composite index's case analysis demonstrates that, in terms of interval root mean squared error (IRMSE), interval mean absolute error (IMAE), interval mean absolute percentage error (IMAPE), and interval mean squared percentage error (IMSPE), the proposed model's predictive performance improvements over the best-performing single prediction model are 29.33%, 25.72%, 26.10%, and 28.86%, respectively. At the same time, the theoretical properties of the model are verified in the results of the case analysis, and the model's convergence is reflected in sensitivity analysis. Through extensive model comparisons, it is observed that the model proposed in this paper exhibits strong generalization, without specific limitations on data size or feature count. It demonstrates good aggregation prediction performance for interval data. Moreover, it is applicable to various fields, including finance, environment, and others.

Improving vertical solar still performance for efficient Desalination: Investigating the influence of Wick, condensate plate and device dimensions

Access to sustainable energy sources and clean water remain two of the most significant challenges in the modern world. Fortunately, desalination can be achieved using a variety of renewable energy sources, with solar energy being a notable option. There has been a noticeable increase in interest in the vertical solar still (VSS) as a popular desalination technique in recent years. The performance of VSS is examined in this article in relation to a several variables, including the wick, the wettability of the condensate plate surface, and device dimensions. According to the study's findings, selecting an appropriate wick is essential for ensuring optimum water distribution on the heat absorber surface, thereby enhancing the effectiveness of the Vertical Solar Still (VSS). In comparison to gauze, cotton emerges as a superior wick, contributing to a noteworthy 5.1 % improvement in device performance. The performance of the VSS is significantly influenced by the contact angle of the condensate surface plate, a glass plate (contact angle of 5) yields 34 % more freshwater than a superhydrophobic plate (contact angle of 140). This study also shows that the dimensions of the vertical solar still (VSS) are significantly impactby the properties of the condensate plate and wick. Specifically, the most efficient VSS performance is attained when the dimensions are set at 32 cm × 30 cm.

Study on Microwave Absorption Performance of Submicron SrFe12O19 Particles Prepared by Hydrothermal Method

: Strontium hexaferrite (SrFe12O19) materials were synthesized through the hydrothermal method with conditions at R= (Fe3++ Sr2+)/(OH ̶) ratio of 2 to 4 (R ratio of 1/2, 1/3 and 1/4 were, respectively abbreviated to M2, M3 and M4). R has been examined by employing Fe(NO3)3 and Sr(NO3)2 as the initial materials. The effects of R ratio on the structure, particle size, magnetic and microwave-absorbing properties of the samples were studied. The coercivity and saturation magnetization of M4 sample reach 5.2 kOe and 58.7 emu/g, respectively. The optimal reflection loss of the M4 at the thickness of 1.5 mm reached a value as large as - 48.27 dB in the frequency range of 2 to 18 GHz. The improvement in microwave absorption performance of M4 sample compared to M2 and M3 was mainly resulted from the increased magnetic properties with increasing R ratio. 
 

Compact Instruction Set Extensions for Dilithium

Post-quantum cryptography is considered to provide security against both traditional and quantum computer attacks. Dilithium is a digital signature algorithm that derives its security from the challenge of finding short vectors in lattices. It has been selected as one of the standardizations in the NIST post-quantum cryptography project. Hardware-software co-design is a commonly adopted implementation strategy to address various implementation challenges, including limited resources, high performance, and flexibility requirements. In this study, we investigate using compact instruction set extensions (ISEs) for Dilithium, aiming to improve software efficiency with low hardware overheads. To begin with, we propose tightly coupled accelerators that are deeply integrated into the RISC-V processor. These accelerators target the most computationally demanding components in resource-constrained processors, such as polynomial generation, Number Theoretic Transform (NTT), and modular arithmetic. Next, we design a set of custom instructions that seamlessly integrate with the RISC-V base instruction formats, completing the accelerators in a compact manner. Subsequently, we implement our ISEs in a chip design for the Hummingbird E203 core and conduct performance benchmarks for Dilithium utilizing these ISEs. Additionally, we evaluate the resource consumption of the ISEs on FPGA and ASIC technologies. Compared to the reference software implementation on the RISC-V core, our co-design demonstrates a remarkable speedup factor ranging from 6.95 to 9.96. This significant improvement in performance is achieved by incorporating additional hardware resources, specifically, a 35% increase in LUTs, a 14% increase in FFs, 7 additional DSPs, and no additional RAM. Furthermore, compared to the state-of-the-art approach, our work achieves faster speed performance with a reduced circuit cost. Specifically, the usage of additional LUTs, FFs, and RAMs is reduced by 47.53%, 50.43%, and 100%, respectively. On ASIC technology, our approach demonstrates 12, 412 cell counts. Our co-design provides a better tradeoff implementation on speed performance and circuit overheads.

DCEF2-YOLO: Aerial Detection YOLO with Deformable Convolution–Efficient Feature Fusion for Small Target Detection

Deep learning technology for real-time small object detection in aerial images can be used in various industrial environments such as real-time traffic surveillance and military reconnaissance. However, detecting small objects with few pixels and low resolution remains a challenging problem that requires performance improvement. To improve the performance of small object detection, we propose DCEF 2-YOLO. Our proposed method enables efficient real-time small object detection by using a deformable convolution (DFConv) module and an efficient feature fusion structure to maximize the use of the internal feature information of objects. DFConv preserves small object information by preventing the mixing of object information with the background. The optimized feature fusion structure produces high-quality feature maps for efficient real-time small object detection while maximizing the use of limited information. Additionally, modifying the input data processing stage and reducing the detection layer to suit small object detection also contributes to performance improvement. When compared to the performance of the latest YOLO-based models (such as DCN-YOLO and YOLOv7), DCEF 2-YOLO outperforms them, with a mAP of +6.1% on the DOTA-v1.0 test set, +0.3% on the NWPU VHR-10 test set, and +1.5% on the VEDAI512 test set. Furthermore, it has a fast processing speed of 120.48 FPS with an RTX3090 for 512 × 512 images, making it suitable for real-time small object detection tasks.

O160: Novel Monitoring of General Surgeons' Intraoperative Cognitive Load following Surgical Sabermetrics Principles

Abstract Introduction Surgeons strive to continually enhance performance. Currently, assessment relies on subjective, retrospective methods-hindering real-time improvement and error prevention. Learning from sports, surgeons can utilise surgical sabermetrics for data-driven analysis. Cognitive load(CogL) impacts performance, with overload impairing skills and judgement. Sabermetrics can be leveraged to measure CogL objectively(e.g.Electrodermal activity(EDA)) and subjectively. This study aimed to combine these measures and explore their relationship with surgeon factors and patient outcomes. Methods Fourteen surgeons conducting 58 benign surgery cases participated. Surgeons’ intraoperative EDA was captured using a digital sensor. Subjective data collected included (i)Surgery Task-Load-Index(SURG-TLX)), (ii)mood, (iii)sleep-disturbance and (iv)risk-aversion. EDA underwent decomposition into (a)tonic (background autonomic arousal) and (b) phasic(response to events) components. Patient outcomes included (i)30-day morbidity and (ii)case difficulty. Results Overall, mean phasic-EDA was 0.038μS and mean tonic-EDA was 0.483μS(ranges 0-1:1 indicating highest CogL). Mean SURG-TLX was 47.128(/120, where 120 is the highest workload). Negative mood related to changes in tonic-EDA(p<0.05,F=11.84). EDA climbed throughout each case (Tonic:p<0.05,F=2.035;Phasic:p<0.05,F=1.626). SURG-TLX were weakly correlated with sleep(p<0.05,ϱ=-0.273) and moderately correlated with case duration(p<0.05, ϱ=0.598), but not EDA(p>0.50). Discussion This study is the first in the UK to report combined CogL measures gathered during live surgery. Awareness of factors that affect CogL crucially allows for the management of factors to prevent overload and to inform targeted interventions for performance improvement. Using digital sensors to measure CogL provides opportunity for automated, real-time performance assessment, advanced feedback, and behavioural modifications.

Mesosilicate materials for environmental applications: Adsorption separation analysis via hybrid computational and machine learning tools

In this extensive analysis, we delve into the utilization of advanced hybrid model for molecular separation description via mesoporous materials. The process considered here is adsorption which is utilized for molecular separation in liquid solution. Data for adsorption are obtained in terms of solute concentration by solving mass transfer equations numerically. Then, the data is used for training and validation of several machine learning techniques to establish the hybrid model. For machine learning model, several methods are utilized including three robust tree-based ensemble models, namely Gradient Boosting (GB), Random Forest (RF), and Extra Trees (ET). The methos were utilized to tackle the challenging task of predicting chemical concentrations in a dataset comprising an array of over 19,000 data. These data points are characterized by input variables representing coordinates (x and y), while the output variable encapsulates chemical concentration (C). By subjecting these models to a meticulous optimization process, we achieved substantial enhancements in their predictive capabilities. RF emerged as the standout performer, boasting a remarkable R2 score of 0.99902. ET and GB also demonstrated commendable performance improvements, underscoring the versatility of tree-based ensemble models in delivering precise and reliable predictions for chemical concentrations within intricate and multifaceted datasets.

Experimental and Computational Study of the Thermal Insulation Properties of Hollow Glass Microspheres

Hollow glass microspheres (HGM) are expected to be used in cryogenic liquid storage tanks as an insulating material with better insulation performance and lower maintenance cost instead of perlite. Obtaining the effective thermal conductivity of HGM can help to improve the insulation performance of cryogenic tanks. Herein, a test system based on the transient plane source method is designed and constructed, and the effective thermal conductivity of six models of HGMs from four suppliers at 200–300 K is measured. The results show that under atmosphere pressure of 200 K, the effective thermal conductivity of HGM is only 25.87–35.84 mW m−1 K−1. The effective thermal conductivity at room and low temperatures is calculated using the derived system of equations for integrated thermal conductivity and radiative heat transfer, with the error of no more than 13% from the current experimental results. The integrated thermal conductivity is still dominant in low pressure and low temperature, and the reduction of the true density contributes to the improvement of the HGM insulation performance.

Aligning total quality management, continuous improvement for process performance: An empirical review

The fundamental goal of this research is to develop a conceptual framework for performance, continuous process improvement, and total quality management. Exploring the literature on total quality management and other variables in connection to various relationships has been the subject of recent investigations. This paper provides an introduction of total quality management, explains how it transforms into a sophisticated strategy employed in modern businesses, and discusses the literature-based research that has been done in this field. This study suggests that total quality management, a concept popularized by Deming, be investigated in a range of industries in order to enhance product quality, reduce waste, and increase productivity. This study is an integrative evaluation of the prior research on the idea of overall quality management and its connection. Future researchers are urged to make use of descriptive survey and empirical technique to generalize its findings. To reflect comprehensive quality management, continuous improvement, and process performance, this study created a conceptual framework and model. The author created a summary table to display earlier research on the subject, including authors, topics, countries, methodologies, and findings. In light of the aforementioned, the researcher filled a knowledge gap.

Corporate Governance Practices and Performance of Deposit Money Banks in Nigeria

Performance of deposit money banks in Nigeria. The specific objective of the study was to critically appraise the relationship between size of board of directors, composition of board members, frequency of board meetings and return on assets of deposit money banks in Nigeria. The data were sourced through secondary sources from annual reports and accounts of sampled deposit money banks in Nigeria. The stated Null Hypotheses were tested through data analysis by using the correlation analysis as analytical tool. The research findings reveal that board size has a positive and strong relationship with return on assets while board composition has a positive but moderately strong association with return on assets. Furthermore, frequency of board meetings has a negative and very weak relationship with return on assets of deposit money banks in Nigeria. The implication of the findings is that increased board size could result in the improvement of financial performance of deposit money banks. The research found that such increase in number of members of the board will generate the desired outcome if it centers on independent nonexecutive directors with wealth of corporate governance experience, sound and profitable contacts, good and relevant education. The negative relationship with frequency of board meetings implies that banks should begin to trim down on number of board meetings as research has found that frequent meetings signal a crisis or distress situation with perceptions of going concern issues and bank failure. The study recommends that new independent non-executive professionals with critical governance and management attributes could be introduced into the board to improve the quality of decisions, earnings and general performance. Frequency of Board Meetings should be reduced to save cost and time while virtual meetings should be called more often than physical meetings as distance is no longer a barrier.

Analyzing commercial grape farm efficiency in Armavir region (Armenia) by using two-stage empirical approach

The purpose of this paper is to provide an empirical assessment of the economic efficiency of grape-producing farms in Armenia. Upon reviewing various field-related studies the frontier analysis was singled out as a methodological base of this study. More specifically two-stage empirical analysis was performed, which includes the measurement of efficiency levels of grape farms by implementing the DEA technique and then assessing the determinants of obtained efficiency scores by performing Tobit modeling. To obtain necessary data, 365 grape farms from the Armavir region were surveyed. The main findings of this paper suggest that the average efficiency score for grape farms is 0.72, and there is room for improvement in the economic performance of farms with 28%. The main determinants of farm efficiency were cultivated grape varieties, farm size, and selling prices of grapes. The obtained results mainly support the findings of similar studies carried out for various viticulture regions across the world. This study provides some methodology bases for further expansion of similar studies both in terms of including the other Armenian viticulture regions and different years to explore the changes in the efficiency of grape farms over time. This article provides a base of knowledge for policymakers, scholars, researchers, investors, and credit companies for their decision-making processes and other purposes.

Performance Improvement of True Time Delay Based Centralized Beamforming Control With the Modulation Instability Phenomenon for Wireless Array Antennas

Performance Improvement of True Time Delay Based Centralized Beamforming Control With the Modulation Instability Phenomenon for Wireless Array Antennas

Intelligent approach for performance improvement of stepper actuator devoted to a medical and clinical application

Intelligent approach for performance improvement of stepper actuator devoted to a medical and clinical application