Complex Search Research Articles

The yin yang of AI: Exploring how commercial and non-commercial orientations shape machine learning innovation

The scale of the potential implications of machine learning (ML) has prompted discussions on the issues of corporate control and technological openness. However, how commercial and non-commercially oriented organisations each contribute to ML progress remains an open question. This study uses the recombinant innovation perspective as a lens to explore recombinant patterns across projects in an open source software (OSS) environment – where a great deal of ML innovation occurs – and assess how commercial orientation influences such patterns. It builds on a unique dataset containing data on 28,443 OSS projects, their code dependencies and the organisations owning them. Exploratory analyses reveal that ML projects combine larger and more diverse components, and produce more atypical combinations in shorter timeframes than other OSS projects, and that both company and non-company owned ML projects contribute to such recombinant atypicality. Regression analyses indicate that company owned ML projects tend to rely more on distant combinations of technical knowledge, whereas non-company owned ML projects tend to produce more novel combinations of application ideas. By extending the theories of recombinant innovation and motivation in OS innovation into a new setting – ML technology, this study contributes to both literatures by confirming that the link between distant recombination and innovation still holds in contexts characterised by complex search spaces, and by suggesting complementarities between commercial and non-commercial orientations in OSS environments rich in knowledge diversity and recombinant activity.

Fast Coding Unit Partitioning Algorithm for Video Coding Standard Based on Block Segmentation and Block Connection Structure and CNN

The recently introduced Video Coding Standard, VVC, presents a novel Quadtree plus Nested Multi-Type Tree (QTMTT) block structure. This structure enables a more flexible block partition and demonstrates enhanced compression performance compared to its predecessor, HEVC. However, The introduction of the new structure has led to a more complex partition search process, resulting in a considerable increase in time complexity. The QTMTT structure yields diverse Coding Unit (CU) block sizes, posing challenges for CNN model inference. In this study, we propose a representation structure termed Block Segmentation and Block Connection (BSC), rooted in texture features. This ensures that partial CU blocks are uniformly represented in size. To address different-sized CUs, various levels of CNN models are designed for prediction. Moreover, we introduce a post-processing method and a multi-thresholding scheme to further mitigate errors introduced by CNNs. This allows for flexible and adjustable acceleration, achieving a trade-off between coding time complexity and performance. Experimental results indicate that, in comparison to VTM-10.0, our “Fast” scheme reduces the average complexity by 57.14% with a 1.86% increase in BDBR. Meanwhile, the “Moderate” scheme reduces average complexity by 50.14% with only a 1.39% increase in BDBR.

Faster Person Re-Identification: One-Shot-Filter and Coarse-to-Fine Search.

Fast person re-identification (ReID) aims to search person images quickly and accurately. The main idea of recent fast ReID methods is the hashing algorithm, which learns compact binary codes and performs fast Hamming distance and counting sort. However, a very long code is needed for high accuracy (e.g., 2048), which compromises search speed. In this work, we introduce a new solution for fast ReID by formulating a novel Coarse-to-Fine (CtF) hashing code search strategy, which complementarily uses short and long codes, achieving both faster speed and better accuracy. It uses shorter codes to coarsely rank broad matching similarities and longer codes to refine only a few top candidates for more accurate instance ReID. Specifically, we design an All-in-One (AiO) module together with a Distance Threshold Optimization (DTO) algorithm. In AiO, we simultaneously learn and enhance multiple codes of different lengths in a single model. It learns multiple codes in a pyramid structure, and encourage shorter codes to mimic longer codes by self-distillation. DTO solves a complex threshold search problem by a simple optimization process, and the balance between accuracy and speed is easily controlled by a single parameter. It formulates the optimization target as a Fβ score that can be optimised by Gaussian cumulative distribution functions. Besides, we find even short code (e.g., 32) still takes a long time under large-scale gallery due to the O(n) time complexity. To solve the problem, we propose a gallery-size-free latent-attributes-based One-Shot-Filter (OSF) strategy, that is always O(1) time complexity, to quickly filter major easy negative gallery images, Specifically, we design a Latent-Attribute-Learning (LAL) module supervised a Single-Direction-Metric (SDM) Loss. LAL is derived from principal component analysis (PCA) that keeps largest variance using shortest feature vector, meanwhile enabling batch and end-to-end learning. Every logit of a feature vector represents a meaningful attribute. SDM is carefully designed for fine-grained attribute supervision, outperforming common metrics such as Euclidean and Cosine metrics. Experimental results on 2 datasets show that CtF+OSF is not only 2% more accurate but also 5× faster than contemporary hashing ReID methods. Compared with non-hashing ReID methods, CtF is 50× faster with comparable accuracy. OSF further speeds CtF by 2× again and upto 10× in total with almost no accuracy drop.

Understanding the Impact of Animation Technology in Virtual Reality: A Systematic Literature Review

This systematic literature review examines at the impact of animation technology on virtual reality (VR) environments. Using complex search techniques on the Scopus and ERIC databases, 30 articles were chosen, with an emphasis on the keywords "animation," "virtual reality," and "technology." The investigation identifies three major themes: improved user experience, educational integration, and industry innovations. The study investigates how animation innovations contribute to immersion, presence, and involvement in virtual reality contexts. Educational Integration studies the use of animation technology in training simulations, virtual classrooms, and skill development. Industry Innovations evaluates the impact of animation technology on various industries, highlighting developing trends and potential for innovation. This study, through a thorough analysis of the literature, gives significant insights into the varied relationship between animation technology and VR, shedding light on its importance for user engagement, educational applications, and industrial advancement. Keywords include animation, virtual reality, technology, enhanced user experience, educational integration, and industry innovations.

Depth-Optimized Quantum Circuits for ASCON: AEAD and HASH

Quantum computing advancements pose security challenges for cryptography. Specifically, Grover’s search algorithm affects the reduction in the search complexity of symmetric-key encryption and hash functions. Recent efforts have been made to estimate the complexity of Grover’s search and evaluate post-quantum security. In this paper, we propose a depth-optimized quantum circuit implementation for ASCON, including both symmetric-key encryption and hashing algorithms, as a part of the lightweight cryptography standardization by NIST (National Institute of Standards and Technology). As far as we know, this is the first implementation of a quantum circuit for the ASCON AEAD (Authenticated Encryption with Associated Data) scheme, which is a symmetric-key algorithm. Also, our quantum circuit implementation of the ASCON-HASH achieves a reduction of more than 88.9% in the Toffoli depth and more than 80.5% in the full depth compared to the previous work. As per our understanding, the most effective strategy against Grover’s search involves minimizing the depth of the quantum circuit for the target cipher. We showcase the optimal Grover’s search cost for ASCON and introduce a proposed quantum circuit optimized for depth. Furthermore, we utilize the estimated cost to evaluate post-quantum security strength of ASCON, employing the relevant evaluation criteria and the latest advancements in research.

Revisiting face detection: Supercharging Viola-Jones with particle swarm optimization for enhanced performance

In recent years, face detection has emerged as a prominent research field within Computer Vision (CV) and Deep Learning. Detecting faces in images and video sequences remains a challenging task due to various factors such as pose variation, varying illumination, occlusion, and scale differences. Despite the development of numerous face detection algorithms in deep learning, the Viola-Jones algorithm, with its simple yet effective approach, continues to be widely used in real-time camera applications. The conventional Viola-Jones algorithm employs AdaBoost for classifying faces in images and videos. The challenge lies in working with cluttered real-time facial images. AdaBoost needs to search through all possible thresholds for all samples to find the minimum training error when receiving features from Haar-like detectors. Therefore, this exhaustive search consumes significant time to discover the best threshold values and optimize feature selection to build an efficient classifier for face detection. In this paper, we propose enhancing the conventional Viola-Jones algorithm by incorporating Particle Swarm Optimization (PSO) to improve its predictive accuracy, particularly in complex face images. We leverage PSO in two key areas within the Viola-Jones framework. Firstly, PSO is employed to dynamically select optimal threshold values for feature selection, thereby improving computational efficiency. Secondly, we adapt the feature selection process using AdaBoost within the Viola-Jones algorithm, integrating PSO to identify the most discriminative features for constructing a robust classifier. Our approach significantly reduces the feature selection process time and search complexity compared to the traditional algorithm, particularly in challenging environments. We evaluated our proposed method on a comprehensive face detection benchmark dataset, achieving impressive results, including an average true positive rate of 98.73% and a 2.1% higher average prediction accuracy when compared against both the conventional Viola-Jones approach and contemporary state-of-the-art methods.

Global convergence analysis of Caputo fractional Whittaker method with real world applications

The present article deals with the effect of convexity in the study of the well-known Whittaker iterative method, because an iterative method converges to a unique solution \(t^*\) of the nonlinear equation \(\psi(t)=0\) faster when the function's convexity is smaller. Indeed, fractional iterative methods are a simple way to learn more about the dynamic properties of iterative methods, i.e., for an initial guess, the sequence generated by the iterative method converges to a fixed point or diverges. Often, for a complex root search of nonlinear equations, the selective real initial guess fails to converge, which can be overcome by the fractional iterative methods. So, we have studied a Caputo fractional double convex acceleration Whittaker's method (CFDCAWM) of order at least (\(1+2\zeta\)) and its global convergence in broad ways. Also, the faster convergent CFDCAWM method provides better results than the existing Caputo fractional Newton method (CFNM), which has (\(1+\zeta\)) order of convergence. Moreover, we have applied both fractional methods to solve the nonlinear equations that arise from different real-life problems.

Enhancement of Genetic Algorithm by J.Zhang Applied to Tour Planning

Following widespread lockdowns, there has been a notable increase in people's desire to travel, leading to longer and more frequent trips. This trend has created a demand for customized itineraries and tour planning. Unfortunately, manual tour planning can be challenging to optimize, time-consuming, and increasingly complex as the number of locations increases. To automate and improve tour planning, optimization methods can be used, as they leverage algorithms to find efficient routes. The Genetic Algorithm (GA), an algorithm that mimics the course of natural evolution, is adept at navigating complex search spaces and finding optimal solutions, making it suitable for solving tour planning challenges. Building upon the work of J. Zhang (2021), this study aims to improve the performance of GA by enhancing the diversity of the population, removing redundant nodes, and reducing the execution time. Two simulators were created, one for each algorithm, to test their performance. The researchers conducted tests on both the existing and enhanced algorithms. This involved the utilization of several test data that contains coordinates of several cities in the Philippines. Based on the results, the enhanced algorithm showed better results compared to the existing algorithm. In conclusion, the enhanced algorithm performed better than the exiting algorithm

Focus of attention in musical learning and music performance: a systematic review and discussion of focus instructions and outcome measures.

The topic of attentional focus (focus of attention, FOA) in musical learning and performance has recently received increasing interest, as the growing number of empirical studies inspired by the established FOA paradigm in sports by Wulf and colleagues in 1998. The current systematical review aims at collecting, abstracting, and categorizing relevant data to show which kinds of FOA instructions were applied in experimental designs and what kinds of dependent variables were used to measure the effects of FOA instruction on musical performance. The three main inclusion criteria in the selection process were experimental design, detailed descriptions of FOA instructions, and outcome measures (OMs). A systematic search was conducted with a complex search term in four scientific databases in March 2023. For presenting and synthesizing results, we used data collection and an inductive-deductive data categorization. Fifteen studies with a total sample size of 401 participants were included out of 387 records initially identified. We collected 53 different FOA instruction citations from the 15 studies and classified them into 9 FOA subcategories, of which the most applied were bodily focus (21%), sound focus (15%), and visual focus (14%). Selected studies used 63 OMs that were abstracted to 10 different OM categories with expert ratings (27%) and acoustical analysis (22%) as the most applied dependent variables. Data categorization and abstraction of additional study information show multiple combinations of FOA instructions, OMs, participants' instruments and expertise, and musical tasks. Finally, studies show no consistent results of superiority of either external or internal or otherwise different FOA considering positive effects on musical performance. Limitations of the review lie in the small study sample, possible criticism of applied eligibility criteria, and subjectivity of data categorization. We propose a research agenda with a more exploratory approach that comprehensively and qualitatively examines the dimensions of musical goals to create a database that could provide a foundation for developing a music-specific FOA model.

MorphVAE: Advancing Morphological Design of Voxel-Based Soft Robots with Variational Autoencoders

Soft robot design is an intricate field with unique challenges due to its complex and vast search space. In the past literature, evolutionary computation algorithms, including novel probabilistic generative models (PGMs), have shown potential in this realm. However, these methods are sample inefficient and predominantly focus on rigid robots in locomotion tasks, which limit their performance and application in robot design automation. In this work, we propose MorphVAE, an innovative PGM that incorporates a multi-task training scheme and a meticulously crafted sampling technique termed ``continuous natural selection'', aimed at bolstering sample efficiency. This method empowers us to gain insights from assessed samples across diverse tasks and temporal evolutionary stages, while simultaneously maintaining a delicate balance between optimization efficiency and biodiversity. Through extensive experiments in various locomotion and manipulation tasks, we substantiate the efficiency of MorphVAE in generating high-performing and diverse designs, surpassing the performance of competitive baselines.

Enhancing Breast Cancer Classification: A Deep Learning Approach with Outlier Detection With Egret Swarm Optimization Algorithm and Feature Selection Integration – A Systematic Review

Cancer, the second leading cause of death worldwide, encompasses a spectrum of diseases characterized by uncontrolled cell proliferation. Effective therapy relies on screening, early diagnosis, and recurrence prediction. Timely identification of lesions is critical for improving patient survival rates, and accurate classification of benign cancers can prevent unnecessary treatment. Breast cancer (BC) classification into benign or malignant categories is a subject of extensive review. This review discusses a novel approach using deep neural networks (DNN) and genetic algorithms (GA) based on Egret Swarm Optimization (ESO) to address this challenge. Initially, an Enhanced Linear Discriminant Analysis (ILDA) technique is applied for data dimensionality reduction. Subsequently, DNN controls outlier detection to identify anomalies in the cancer dataset. GA is then employed for feature selection, efficiently locating optimal features in complex search spaces, crucial for cancer diagnosis models, especially non-invasive ones. Finally, ESO algorithm, integrating the aggressive and sit-and-wait tactics of Great Egret and Snowy Egret, respectively, is utilized to classify malignancy. In ths review, The proposed method is evaluated using BC datasets, with trials conducted before and after outlier removal. Experimental results demonstrate the effectiveness and accuracy of the proposed strategy. The findings of this study hold promise for enhancing BC diagnosis and guiding patient treatment decisions.

Neural architecture search via progressive partial connection with attention mechanism

Differentiable architecture search requires a larger computational consumption during architecture search, and there exists the depth gap problem under deeper network architecture. In this paper, we propose an attention-based progressive partially connected neural architecture search method (PPCAtt-NAS) to address these two issues. First, we introduce a progressive search strategy in the architecture search phase, build up the sophistication of the architecture gradually and perform path-level pruning in stages to bridge the depth gap. Second, we adopt a partial search scheme that performs channel-level partial sampling of the network architecture to further reduce the computational complexity of the architecture search. In addition, an attention mechanism is devised to improve the architecture search capability by enhancing the relevance between the feature channels. Finally, we conduct extensive comparison experiments with state-of-the-art methods on several public datasets, and our method is able to present higher architecture performance.

Bayesian hybrid-kernel machine-learning-assisted sensitivity analysis and sensitivity-relevant inverse modeling for groundwater DNAPL contamination

Accurate source characterization and transport parameter estimation is important when seeking to predict the spatiotemporal distribution of dense non-aqueous phase liquid (DNAPL) contaminants in groundwater. However, this is a complex multimodal search problem prone to equifinality and premature convergence, which leads to considerable error. To address this, a sensitivity-relevant dynamic swarm intelligence (SRD-SI) algorithm embedded in a homotopy-variation mechanism is proposed in the present study to rationally balance the inversion processes of sensitivity-varied source characteristics and DNAPL transport parameters. In this approach, global optima are progressively approached in conjunction with the homotopy variation of the search space. Furthermore, to avoid computationally expensive numerical model repetition during the sensitivity analysis and inverse iterations, a Bayesian-based optimization framework that combines multiple kernel functions in a kernel extreme learning machine (KELM) model is designed considering the complex site conditions and statistical characteristics of the input variables, thus creating the Bayesian hybrid KELM (BHK-ELM) model for the reliable surrogate modeling of numerical DNAPL-transport simulations. The results show that the BHK-ELM model recognizes and effectively reconstructs the complex input − output mapping of the numerical model by increasing the determination coefficient R2 to 0.9988 while improving the computational efficiency approximately 4500-fold. Because source characteristics and boundary conditions are far more sensitive than transport parameters to the contaminant distribution, conventional inverse modeling methods struggle to accurately identify these transport parameters. In contrast, the proposed inverse modeling system combining sensitivity analysis, swarm intelligence, and homotopy variation is more stable and provides significantly more accurate estimations for all unknown variables. Compared with the traditional SI algorithm, the homotopy-variation SRD-SI reduced the maximum inversion relative error from 46.22 % to 9.53 %, while the mean inversion relative error was reduced from 11.09 % to 3.90 %.

Solving polynomial systems over non-fields and applications to modular polynomial factoring

We study the problem of solving a system of m polynomials in n variables over the ring of integers modulo a prime-power pk. The problem over finite fields is well studied in varied parameter settings. For small characteristic p=2, Lokshtanov et al. (SODA'17) initiated the study, for degree d=2 systems, to improve the exhaustive search complexity of O(2n)⋅poly(m,n) to O(20.8765n)⋅poly(m,n); which currently is improved to O(20.6943n)⋅poly(m,n) in Dinur (SODA'21). For large p but constant n, Huang and Wong (FOCS'96) gave a randomized poly(d,m,log⁡p) time algorithm. Note that for growing n, system-solving is known to be intractable even with p=2 and degree d=2.We devise a randomized poly(d,m,log⁡p)-time algorithm to find a root of a given system of m integral polynomials of degrees bounded by d, in n variables, modulo a prime power pk; when n+k is constant. In a way, we extend the efficient algorithm of Huang and Wong (FOCS'96) for system-solving over Galois fields (i.e., characteristic p) to system-solving over Galois rings (i.e., characteristic pk); when k>1 is constant. The challenge here is to find a lift of singularFp-roots (exponentially many); as there is no efficient general way known in algebraic-geometry for resolving singularities.Our algorithm has applications to factoring univariate polynomials over Galois rings. Given f∈Z[x] and a prime-power pk (k≥2), finding factors of fmodpk has a curious state-of-the-art. It is solved for large k by p-adic factoring algorithms (von zur Gathen, Hartlieb, ISSAC'96); but unsolved for small k. In particular, no nontrivial factoring method is known for k≥5 (Dwivedi, Mittal, Saxena, ISSAC'19). One issue is that degree-δ factors of f(x)modpk could be exponentially many, as soon as k≥2. We give the first randomized poly(deg⁡(f),log⁡p)-time algorithm to find a degree-δ factor of f(x)modpk, when k+δ is constant. Our method has potential application in algebraic coding theory. In particular, extending algebraic geometric and Reed-Solomon codes to Galois rings could enable new and improved bounds on their underlying efficiency parameters.

Ruthenium(II) Complexes as Potential Apoptosis Inducers in Cancer Therapy

Abstract The compound cis-diamminedichloroplatinum(II) (cisplatin) is the most widely used anticancer drug, but due to its serious side effects (including gastrointestinal symptoms, renal tubular injury, neuromuscular complications, and ototoxicity), clinical applications of cisplatin are limited. Therefore, these limitations have provided an encouragement for further research into other transition metal complexes, with an aim to overcome the disadvantages related with cisplatin therapy. In the search for effective complexes that can be targeted against tumor cells, many research groups synthesized various ruthenium(II) complexes with different ligands. Also, newly synthesized ruthenium(II) complexes showed selective anticancer activity against different types of cancer cells. Activity of ruthenium(II) complexes in some cases was even higher than that of cisplatin against the same cells. Precise mechanism of action of ruthenium(II) complexes is not fully understood. The different examples mentioned in this review showed that ruthenium(II) complexes decreased viability of cancer cells by induction of apoptosis and/or by cell cycle arrest which implies their different mechanism of action against different types of cancer cells.

Neurovascular coupling is altered in women who have a history of brain injury from intimate partner violence: a preliminary study.

Intimate partner violence (IPV) is a global health crisis with 30% of women over the age of 15 experiencing at least one event in their lifetime. Brain injury (BI) due to head impacts and/or strangulation is a common but understudied part of this experience. Previous research has shown BI from other injury mechanisms can disrupt neurovascular coupling (NVC). To gain further insight into whether similar changes occur in this population, we assessed NVC responses in women with a history of IPV-BI. NVC responses were measured for the middle and posterior cerebral arteries (MCA, PCA) using transcranial Doppler ultrasound while participants performed a complex visual search task. The lifetime history of previous exposure to IPV-BI was captured using the Brain Injury Severity Assessment (BISA) along with measures of post-traumatic stress disorder (PTSD), anxiety, depression, substance use, and demographic information. Initial analyses of NVC metrics were completed comparing participants who scored low vs. high on the BISA or did or did not experience non-fatal strangulation followed by a stepwise multiple regression to examine the impact of PTSD, anxiety, and depression on the relationship between the NVC metrics and IPV-BI. Baseline and peak cerebral blood velocity were higher and the percentage increase was lower in the PCA in the low compared to the high BISA group whereas no differences between the groups were apparent in the MCA. In addition, those participants who had been strangled had a lower initial slope and area under the curve in the PCA than those who had not experienced strangulation. Finally, the stepwise multiple regression demonstrated the percentage increase in the PCA was significantly related to the BISA score and both depression and anxiety significantly contributed to different components of the NVC response. This preliminary study demonstrated that a lifetime history of IPV-BI leads to subtle but significant disruptions to NVC responses which are modulated by comorbid depression and anxiety. Future studies should examine cerebrovascular function at the acute and subacute stages after IPV episodes to shed additional light on this experience and its outcomes.

Anomaly detection in time-series data using evolutionary neural architecture search with non-differentiable functions

Deep neural networks have become the benchmark in diverse fields such as energy consumption forecasting, speech recognition, and anomaly detection, owing to their ability to efficiently process and analyse data. However, they face challenges in managing the complexity and variability in time series data, often leading to increased model complexity and prolonged search duration during parameter tuning. This paper proposes a novel anomaly detection approach through evolutionary neural architecture search (AD-ENAS), which is specifically designed for time series data. The proposed approach focuses on the search for the optimal and minimal neural network architecture. The AD-ENAS method consists of two main phases: architecture evolution and weight adjustment. The architecture evolution phase highlights the importance of neural network architecture by evaluating the fitness of each network agent using shared weight values. Subsequently, the convolutional matrix adaptation technique is used in the next phase for optimal weight adjustment of the neural network. The proposed AD-ENAS method operates without relying on differentiable functions, thus expanding the scope of neural network design beyond traditional backpropagation-based approaches. Various non-differentiable loss functions are explored to facilitate effective architecture search and weight adjustment. Comparative experiments are conducted with five baseline anomaly detection methods on three well-known datasets from reputable sources such as NASA SMAP, NASA MSL and Yahoo S5-A1. The results demonstrate that the AD-ENAS approach effectively evolves neural network architectures, outperforming baseline methods with F1 scores across the three datasets (MSL: 0.942, SMAP: 0.961, Yahoo S5-A1: 0.988) with non-differentiable loss functions, showcasing its efficacy in detecting anomalies in time series data.

Хронічний коліт у дітей. Проблеми диференціальної діагностики

During the last few decades, there has been observed a "rejuvenation" of the inflammatory bowel diseases (IBD) onset in childhood. Very often the problem is narrowed down to the differential diagnosis of Crohn's disease and ulcerative colitis, which are related to IBD, although in reality the spectrum of chronic IBD is much wider. A diagnosed case of microscopic (collagenous) colitis confirms the need for a wide diagnostic search in patients with IBD. Purpose - to describe differential diagnostic features of collagenous colitis and collagenous gastritis in young children. Clinical case. Observational data on a patient aged 1 year and 7 months with a special form of colitis (collagenous colitis in combination with collagenous gastritis) are presented. The peculiarities of this rare disease course in the examined patient, in whom the pathology debuted with gastric ulcers with hematemesis, anemia and diarrhea syndrome, were analyzed. The differential search complex stages are discussed in detail, the results of laboratory, instrumental and pathomorphological research methods are described. Information is provided on the treatment prospects of patients with a gastrointestinal tract collagenous pathological process. There is a justified need to take into account rare causes of digestive tract inflammation when observing young patients with a clinical picture similar to the IBD debut. The research was carried out in accordance with the principles of the Declaration of Helsinki. The informed consent of the child's parents was obtained for conducting the studies. No conflict of interests was declared by the authors.

Ranking‐based architecture generation for surrogate‐assisted neural architecture search

AbstractArchitectures generation optimization has been received a lot of attention in neural architecture search (NAS) since its efficiency in generating architecture. By learning the architecture representation through unsupervised learning and constructing a latent space, the prediction process of predictors is simplified, leading to improved efficiency in architecture search. However, searching for architectures with top performance in complex and large NAS search spaces remains challenging. In this paper, an approach that combined a ranker and generative model is proposed to address this challenge through regularizing the latent space and identifying architectures with top rankings. We introduce the ranking error to gradually regulate the training of the generative model, making it easier to identify architecture representations in the latent space. Additionally, a surrogate‐assisted evolutionary search method that utilized neural network Bayesian optimization is proposed to efficiently explore promising architectures in the latent space. We demonstrate the benefits of our approach in optimizing architectures with top rankings, and our method outperforms state‐of‐the‐art techniques on various NAS benchmarks. The code is available at https://github.com/outofstyle/RAGS‐NAS.

EVALUASI USABILITAS SISTEM MENGGUNAKAN METODE SYSTEM USABILITY SCALE (SUS) PADA APLIKASI AKHLAQU DENGAN PENERAPAN TEKNIK INDEXING MONG

The Akhlaqu Application is a web-based Islamic learning platform used by Al-Falah Integrated Islamic Elementary School to facilitate the recording and reporting of students’ activities in developing their memorization. This research aims to assess the level of user satisfaction with the Akhlaqu Application using the System Usability Scale (SUS). The results showed a high level of usability, measured by a System Usability Scale (SUS) score of 89.7, indicating a “very good” quality application that meets high usability standards. In addition, the implementation of text indexes on MongoDB has a positive impact on the speed of transaction data search, especially on complex data searches, improving the efficiency and responsiveness of the application. The implications of this research involve recommendations to continue to improve the usability and performance of the Akhlaqu Application, by maintaining the use of MongoDB index techniques as a reliable solution for optimizing the database. The study concludes that the Akhlaqu Application is an effective tool for managing students’ memorization and recommends further improvements to enhance its usability and performance.