Size Of Search Space Research Articles

Exploring Locality Sensitive Hashing as a Blocking Method for Large-Scale Administrative Datasets

ObjectivesLinking large-scale datasets is challenging due to the computational power required. This research explores using Locality-Sensitive-Hashing (LSH) as a novel blocking method to facilitate linking large administrative data. LSH uses an efficient technique of hashing data while preserving similarity, reducing the search space and processing power required to find links. ApproachSamples were taken from a gold-standard dataset and blocked using LSH. Various parameters were tested to establish LSH’s optimal performance at retrieving linked pairs and reducing the search space size. ResultsTesting the method on small datasets gave promising results, with the LSH method creating ~9,000 candidate pairs. This is a improvement over our traditional blocking method or cartesian product, which created ~70,000 and 23.4 million candidate pairs, respectively. We have therefore shown that LSH can significantly reduce the search-space size. A further advantage of the method is its capability in handling alternative variables such as those that may be present in longitudinal or composite data, without needing to manually anticipate the different combinations of variables which may occur. Variable formats to simulate agreement weighting were also considered, with encouraging results. Conclusion and ImplicationsOur research shows that LSH can be used to drastically reduce the search space when blocking for data linkage This suggests developing LSH as a blocking method could result in more effective blocking, quicker linkage, and no loss of quality compared to traditional methods. However, further research on increasing the method’s scale is necessary.

Sequential node search for faster neural architecture search

Neural Architecture Search (NAS) has progressed significantly by reducing search costs since it was first introduced. The faster NAS algorithms use cell-type architectures. The cells are made up of nodes, and typically, all the node operations are searched together. This paper proposes a novel algorithm that searches the nodes of the cell type NAS independently of one another. This is done in a sequential manner, hence named SeQuential Neural Architecture Search (SQNAS). It is hypothesized that the hierarchical nature of nodes’ feature tensors is the proper guide to organize the sequence of the nodes. SQNAS reduces the size of the super graph search space at any instant of search, thus reducing the search cost considerably. The independence assumption is justified considering the competitive performance obtained. SQNAS search is also done in cycles of several restarts for better regularization. With a single NVIDIA GTX 1080Ti GPU, the search is completed in 6 h for the image classification task, obtaining a 2.55% test error on CIFAR10. For the Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) cells searched on smaller datasets, the transferability to deeper models with larger datasets is tested with ImageNet and WikiText-2. Partial channel SQNAS was also implemented, which completed the search in just 1.5 h on the CIFAR10 dataset with competitive test results. This is one of the fastest searches achieved on CIFAR10.

Equilibrium optimizer-based harmony search algorithm with nonlinear dynamic domains and its application to real-world optimization problems

Harmony Search (HS) algorithm is a swarm intelligence algorithm inspired by musical improvisation. Although HS has been applied to various engineering problems, it faces challenges such as getting trapped in local optima, slow convergence speed, and low optimization accuracy when applied to complex problems. To address these issues, this paper proposes an improved version of HS called Equilibrium Optimization-based Harmony Search Algorithm with Nonlinear Dynamic Domains (EO-HS-NDD). EO-HS-NDD integrates multiple leadership-guided strategies from the Equilibrium Optimizer (EO) algorithm, using harmony memory considering disharmony and historical harmony memory, while leveraging the hidden guidance direction information from the Equilibrium Optimizer. Additionally, the algorithm designs a nonlinear dynamic convergence domain to adaptively adjust the search space size and accelerate convergence speed. Furthermore, to balance exploration and exploitation capabilities, appropriate adaptive adjustments are made to Harmony Memory Considering Rate (HMCR) and Pitch Adjustment Rate (PAR). Experimental validation on the CEC2017 test function set demonstrates that EO-HS-NDD outperforms HS and nine other HS variants in terms of robustness, convergence speed, and optimization accuracy. Comparisons with advanced versions of the Differential Evolution (DE) algorithm also indicate that EO-HS-NDD exhibits superior solving capabilities. Moreover, EO-HS-NDD is applied to solve 15 real-world optimization problems from CEC2020 and compared with advanced algorithms from the CEC2020 competition. The experimental results show that EO-HS-NDD performs well in solving real-world optimization problems.

A distributed evolutionary based instance selection algorithm for big data using Apache Spark

Instance selection is an important preprocessing technology in data mining and machine learning. In this paper, we proposed a novel evolutionary based instance selection algorithm for big data. First, we defined a coarse granularity chromosome structure to reduce the size of search space and costs of chromosome operations(recombination and mutation, etc.). Then a stratified evolution strategy was proposed to remove the hyper parameter in classic fitness function and achieve precise control over the reduction ratio of instances. Finally, a sampling-based fitness function was proposed to reduce the time complexity. Experimental results shown that our new algorithm is efficient to complete the instance selection task on data set with millions of instances in minutes-level. The 10-fold cross-validation also proved that the selection results on many datasets have high nearest neighbor classification accuracy.

Reidentification of Participants in Shared Clinical Data Sets: Experimental Study

Background Large curated data sets are required to leverage speech-based tools in health care. These are costly to produce, resulting in increased interest in data sharing. As speech can potentially identify speakers (ie, voiceprints), sharing recordings raises privacy concerns. This is especially relevant when working with patient data protected under the Health Insurance Portability and Accountability Act. Objective We aimed to determine the reidentification risk for speech recordings, without reference to demographics or metadata, in clinical data sets considering both the size of the search space (ie, the number of comparisons that must be considered when reidentifying) and the nature of the speech recording (ie, the type of speech task). Methods Using a state-of-the-art speaker identification model, we modeled an adversarial attack scenario in which an adversary uses a large data set of identified speech (hereafter, the known set) to reidentify as many unknown speakers in a shared data set (hereafter, the unknown set) as possible. We first considered the effect of search space size by attempting reidentification with various sizes of known and unknown sets using VoxCeleb, a data set with recordings of natural, connected speech from >7000 healthy speakers. We then repeated these tests with different types of recordings in each set to examine whether the nature of a speech recording influences reidentification risk. For these tests, we used our clinical data set composed of recordings of elicited speech tasks from 941 speakers. Results We found that the risk was inversely related to the number of comparisons an adversary must consider (ie, the search space), with a positive linear correlation between the number of false acceptances (FAs) and the number of comparisons (r=0.69; P<.001). The true acceptances (TAs) stayed relatively stable, and the ratio between FAs and TAs rose from 0.02 at 1 × 105 comparisons to 1.41 at 6 × 106 comparisons, with a near 1:1 ratio at the midpoint of 3 × 106 comparisons. In effect, risk was high for a small search space but dropped as the search space grew. We also found that the nature of a speech recording influenced reidentification risk, with nonconnected speech (eg, vowel prolongation: FA/TA=98.5; alternating motion rate: FA/TA=8) being harder to identify than connected speech (eg, sentence repetition: FA/TA=0.54) in cross-task conditions. The inverse was mostly true in within-task conditions, with the FA/TA ratio for vowel prolongation and alternating motion rate dropping to 0.39 and 1.17, respectively. Conclusions Our findings suggest that speaker identification models can be used to reidentify participants in specific circumstances, but in practice, the reidentification risk appears small. The variation in risk due to search space size and type of speech task provides actionable recommendations to further increase participant privacy and considerations for policy regarding public release of speech recordings.

AutoMLBench: A comprehensive experimental evaluation of automated machine learning frameworks

With the booming demand for machine learning applications, it has been recognized that the number of knowledgeable data scientists cannot scale with the growing data volumes and application needs in our digital world. In response to this demand, several automated machine learning (AutoML) frameworks have been developed to fill the gap of human expertise by automating the process of building machine learning pipelines. Each framework comes with different heuristics-based design decisions. In this study, we present a comprehensive evaluation and comparison of the performance characteristics of six popular AutoML frameworks, namely, AutoWeka, AutoSKlearn, TPOT, Recipe, ATM and SmartML across 100 data sets from established AutoML benchmark suites. Our experimental evaluation considers different aspects for its comparison, including the performance impact of several design decisions, including time budget, size of search space, meta-learning, and ensemble construction. The results of our study reveal various interesting insights that can significantly guide and impact the design of AutoML frameworks.

Efficient weighted sequential pattern mining

In real-life applications, data mining task involves extracting valuable but hidden information from massive data. How to effectively find out interesting patterns from large databases is a current topic. Sequential pattern mining is the most popular approach in data mining domain. Traditional sequential pattern mining research generally focuses on discovering frequent sequential patterns. However, the account of occurrence times of patterns does not adequately indicate their importance. For instance, frequent patterns (e.g., pencil and eraser) are not profitable, whereas infrequent patterns (e.g., extreme weather) are high-risk. To extract more useful information, researchers study a weighted sequential pattern mining task. In this paper, an efficient algorithm for weighted sequential pattern mining task, called EWSPM, is proposed. Two new strict upper bounds, namely MWEbound and MSRIWbound, are designed based on the concepts of maximum weight estimation (simplified as MWE) and maximum sumation of remaining item weights (simplified as MSRIW), respectively. These upper bounds achieve better pruning effects and reduce the size of search space during the mining process, which significantly shortens execution time. In addition, a database-projection method is employed to optimize memory usage. It addresses potential memory explosion issues in a certain degree. Finally, we also conducted extensive experiments on nine datasets (including real and synthetic). The experimental results demonstrate that the EWSPM algorithm is capable of mining all interesting patterns efficiently, with the smallest size of search space. Additionally, the novel algorithm also exhibits superior performance in terms of execution time and memory consumption.

Bayesian Network Structural Learning Using Adaptive Genetic Algorithm with Varying Population Size

A Bayesian network (BN) is a probabilistic graphical model that can model complex and nonlinear relationships. Its structural learning from data is an NP-hard problem because of its search-space size. One method to perform structural learning is a search and score approach, which uses a search algorithm and structural score. A study comparing 15 algorithms showed that hill climbing (HC) and tabu search (TABU) performed the best overall on the tests. This work performs a deeper analysis of the application of the adaptive genetic algorithm with varying population size (AGAVaPS) on the BN structural learning problem, which a preliminary test showed that it had the potential to perform well on. AGAVaPS is a genetic algorithm that uses the concept of life, where each solution is in the population for a number of iterations. Each individual also has its own mutation rate, and there is a small probability of undergoing mutation twice. Parameter analysis of AGAVaPS in BN structural leaning was performed. Also, AGAVaPS was compared to HC and TABU for six literature datasets considering F1 score, structural Hamming distance (SHD), balanced scoring function (BSF), Bayesian information criterion (BIC), and execution time. HC and TABU performed basically the same for all the tests made. AGAVaPS performed better than the other algorithms for F1 score, SHD, and BIC, showing that it can perform well and is a good choice for BN structural learning.

Augmented ant colony algorithm for virtual drug discovery

Docking is a fundamental problem in computational biology and drug discovery that seeks to predict a ligand’s binding mode and affinity to a target protein. However, the large search space size and the complexity of the underlying physical interactions make docking a challenging task. Here, we review a docking method, based on the ant colony optimization algorithm, that ranks a set of candidate ligands by solving a minimization problem for each ligand individually. In addition, we propose an augmented version that takes into account all energy functions collectively, allowing only one minimization problem to be solved. The results show that our modification outperforms in accuracy and efficiency.

Developing an Intelligent Decision Support System for large-scale smart grid communication network planning

The placement of routers and gateways in wireless communication networks for Smart Grid Advanced Metering Infrastructure (AMI) is a complex problem often addressed in the literature through heuristics, clustering, or linear programming approaches. In large-scale scenarios, the complexity is directly related to the region’s number of smart meters and poles. The terrain profile can introduce signal quality degradation between a smart meter and the routers and gateways, further complicating the problem. This study presents the AIDA-ML method, representing a preliminary step toward developing an Intelligent Decision Support System (IDSS) for an effective positioning strategy. AIDA-ML incorporates a feature engineering-based strategy and utilizes machine learning algorithms to learn from the results computed by a heuristic method called AIDA. The heuristic approach is too time-consuming, relying on external resources like numerous terrain profile API calls and a deep understanding of wireless communication loss models. To implement a machine learning-based and more straightforward approach, a feature engineering process is employed to capture the operational characteristics and results of the analytical method while requiring less processing time. Through experiments conducted with real-world data from 26 cities in the state of Paraná, Brazil, comprising 466,237 smart meters and 352,867 poles, the results obtained using machine learning algorithms suggest that AIDA-ML can ensure the connection coverage of smart meters while meeting the same minimum requirements established for the analytical method. Moreover, AIDA-ML offers the advantage of reducing the processing time by 87.60% and by 96.86% the overall search space size compared to the heuristic approach.

Exploring locality sensitive hashing as a blocking method for large-scale administrative datasets

ObjectivesLinking large-scale datasets is challenging due to the computational power required. This research explores using Locality-Sensitive-Hashing (LSH) as a blocking method to reduce the computational complexity when linking large administrative datasets. LSH hashes similar data into ‘buckets’, thus reducing the search space and processing power required to find links.
 MethodsA gold-standard linked dataset was used during method development. Test datasets were made using samples of gold-standard matches and non-matches, then blocked using LSH.
 Various LSH parameters including shingle length, signature length, band size and number of matching bands were tested. Precision and recall were used to find optimal parameters for identifying good candidate pairs, with 100% recall and >20% precision being desirable.
 Alternative formats for date of birth, postcode and gender variables were tested, with additional characters used to simulate agreement weighting.
 ResultsResults as of spring 2023 are promising, with the caveat that currently only small datasets have been tested. The LSH method with optimal parameters creates ~9,000 candidate pairs whilst maintaining recall of 100% (i.e., all true matches are included in the candidate pairs) and precision of 27.6%. In contrast, our traditional deterministic blocking method using the same variables creates ~70,000 candidate pairs, and a cartesian product creates over 23.4 million candidate pairs. We have therefore shown that LSH can be used to create a significant reduction in the search-space size.
 Furthermore, the method easily handles alternative names, postcodes, etc. that may be present in longitudinal data or composite datasets, with no need to account for different possible combinations of variables.
 ConclusionCurrent research has shown that LSH can be used to drastically reduce the search space when blocking for data linkage. Using variable formatting to prioritise agreement for specific sections e.g., of postcode, has overcome a potential downside of LSH. Further research on variable formatting, parameter optimisation and testing of the method at scale is ongoing.

Convexity Hierarchies in Grid Networks

Several algorithms for path planning in grid networks rely on graph decomposition to reduce the search space size; either by constructing a search data structure on the components, or by using component information for A* guidance. The focus is usually on obtaining components of roughly equal size with few boundary nodes each. In this paper, we consider the problem of splitting a graph into convex components. A convex component is characterized by the property that for all pairs of its members, the shortest path between them is also contained in it. Thus, given a source node, a target node, and a (small) convex component that contains both of them, path planning can be restricted to this component without compromising optimality. We prove that it is NP-hard to find a balanced node separator that splits a given graph into convex components. However, we also present and evaluate heuristics for (hierarchical) convex decomposition of grid networks that perform well across various benchmarks. Moreover, we describe how existing path planning methods can benefit from the computation of convex components. As one main outcome, we show that contraction hierarchies become up to an order of magnitude faster on large grids when the contraction order is derived from a convex graph decomposition.

Comprehensive Approach to Simulating Large Scale Conformational Changes in Biological Systems Utilizing a Path Collective Variable and New Barrier Restraint.

Conformational sampling of complex biomolecules is an emerging frontier in drug discovery. Advances in lab-based structural biology and related computational approaches like AlphaFold have made great strides in obtaining static protein structures for biologically relevant targets. However, biology is in constant motion, and many important biological processes rely on conformationally driven events. Conventional molecular dynamics (MD) simulations run on standard hardware are impractical for many drug design projects, where conformationally driven biological events can take microseconds to milliseconds or longer. An alternative approach is to focus the search on a limited region of conformational space defined by a putative reaction coordinate (i.e., path collective variable). The search space is typically limited by applying restraints, which can be guided by insights about the underlying biological process of interest. The challenge is striking a balance between the degree to which the system is constrained and still allowing for natural motions along the path. A plethora of restraints exist to limit the size of conformational search space, although each has drawbacks when simulating complex biological motions. In this work, we present a three-stage procedure to construct realistic path collective variables (PCVs) and introduce a new kind of barrier restraint that is particularly well suited for complex conformationally driven biological events, such as allosteric modulations and conformational signaling. The PCV presented here is all-atom (as opposed to C-alpha or backbone only) and is derived from all-atom MD trajectory frames. The new restraint relies on a barrier function (specifically, the scaled reciprocal function), which we show is particularly beneficial in the context of molecular dynamics, where near-hard-wall restraints are needed with zero tolerance to restraint violation. We have implemented our PCV and barrier restraint within a hybrid sampling framework that combines well-tempered metadynamics and extended-Lagrangian adaptive biasing force (meta-eABF). We use three particular examples of high pharmaceutical interest to demonstrate the value of this approach: (1) sampling the distance from ubiquitin to a protein of interest within the supramolecular cullin-RING ligase complex, (2) stabilizing the wild-type conformation of the oncogenic mutant JAK2-V617F pseudokinase domain, and (3) inducing an activated state of the stimulator of interferon genes (STING) protein observed upon ligand binding. For examples 2 and 3, we present statistical analysis of meta-eABF free energy estimates and, for each case, code for reproducing this work.

Estimation of a Craig–Bampton equivalent model using a hybrid particle swarm optimization for DCLA purposes

It is common practice in industry to deliver a Craig–Bampton reduced version of the spacecraft’s finite element model to the launch authorities to perform the dynamic coupled load analyses. During the dynamic coupled load analyses, the finite element model of the spacecraft is coupled with the finite element model of the launcher to predict the responses caused by the vibration environment produced during the launch phases (e.g., lift-off, engine start-up, etc.). Before being Craig–Bampton-reduced, the finite element model needs to be correlated and validated against physical test results to guarantee its capability to reproduce the actual behaviour of the physical hardware, which is essential to perform a meaningful dynamic coupled load analyses. Given the complexity and size of the finite element model, the entire process can be very demanding. In this paper, an alternative to the mathematical models usually used in the dynamic coupled load analyses is investigated. The main idea is to synthesize a Craig–Bampton equivalent model directly from the response data of a vibration test experiment. The system identification problem is solved using a particle swarm optimization algorithm, tailored specifically for this type of application, the hybrid PSO and Local Search (hPSO-LS). A mutation operator from genetic algorithm is used to improve the solution and the local search method to refine the final result. The proposed method is numerically tested on a 5-degrees of freedom lumped mass system, with full stiffness and damping matrices. A sensitivity analysis of the search space size is also carried out to investigate its influence on the final results. The hPSO-LS is tested in both noise-free and noisy scenarios.

ENAP: An Efficient Number-Aware Pruning Framework for Design Space Exploration of Approximate Configurations

Approximate computing has emerged as a new computing architecture paradigm that trades off necessary numerical accuracy for performance. Various approximation operation units such as adders and multipliers have been created and provide the basis for improving system efficiency, but it is clear, that a design space exploration (DSE) is needed if improved performance is to be systematically achieved. The challenge is to determine a suitable configuration among approximation units with different error characteristics to ensure a minimization of resources while not exceeding user-defined error constraints. In this paper, we propose the efficient number-aware pruning (ENAP) technique that can compress the search space size. Using common fault-tolerant applications, we demonstrate a compression rate up to 0.0008%, meaning that 99.9992% of invalid designs can remain unsearched. An improved genetic algorithm (GA) is subsequently proposed to improve ENAP, allowing the creation of the optimal configuration in only 2 to 3 iterations, thereby greatly improving search efficiency compared to the initial 9 iterations. We integrate these two approaches into the proposed framework, demonstrating how we can achieve better exploration results compared to state-of-the-artwork.

Quantum Genetic Algorithm for Highly Constrained Optimization Problems

Quantum computing appears as an alternative solution for solving computationally intractable problems. This paper presents a new constrained quantum genetic algorithm designed specifically for identifying the extreme value of a highly constrained optimization problem, where the search space size _database is massive and unsorted_ cannot be handled by the currently available classical or quantum processor, called the highly constrained quantum genetic algorithm (HCQGA). To validate the efficiency of the suggested quantum method, maximizing the energy efficiency with respect to the target user bit rate of an uplink multi-cell massive multiple-input and multiple- output (MIMO) system is considered as an application. Simulation results demonstrate that the proposed HCQGA converges rapidly to the optimum solution compared with its classical benchmark.

Exact Algorithm for Batch Scheduling on Unrelated Machine

In this paper, we propose a new linear algorithm to tackle a specific class of unrelated machine scheduling problem, considered as an important real-life situation, which we called Batch Scheduling on Unrelated Machine (BSUM), where we have to schedule a batch of identical and non-preemptive jobs on unrelated parallel machines. The objective is to minimize the makespan (Cmax) of the whole schedule. For this, a mathematical formulation is made and a lower bound is computed based on the potential properties of the problem in order to reduce the search space size and thus accelerate the algorithm. Another property is also deducted to design our algorithm that solves this problem. The latter is considered as a particular case of RmCmax family problems known as strongly NP-hard, therefore, a polynomial reduction should realize a significant efficiency to treat them. As we will show, Batch BSUM is omnipresent in several kind of applications as manufacturing, transportation, logistic and routing. It is of major importance in several company activities. The problem complexity and the optimality of the algorithm are reported, proven and discussed.

A Computational Complexity Perspective on Segmentation as a Cognitive Subcomputation.

Computational feasibility is a widespread concern that guides the framing and modeling of natural and artificial intelligence. The specification of cognitive system capacities is often shaped by unexamined intuitive assumptions about the search space and complexity of a subcomputation. However, a mistaken intuition might make such initial conceptualizations misleading for what empirical questions appear relevant later on. We undertake here computational-level modeling and complexity analyses of segmentation - a widely hypothesized subcomputation that plays a requisite role in explanations of capacities across domains, such as speech recognition, music cognition, active sensing, event memory, action parsing, and statistical learning - as a case study to show how crucial it is to formally assess these assumptions. We mathematically prove two sets of results regarding computational hardness and search space size that may run counter to intuition, and position their implications with respect to existing views on thesubcapacity.

Offspring regeneration method based on bi-level sampling for large-scale evolutionary multi-objective optimization

In evolutionary multi-objective optimization, the size of search space exponentially expands as the number of decision variables increases, which makes the generation of promising candidate solutions more difficult. For this, in this paper, we propose a bi-level offspring generation architecture, together with a deep offspring sampling method. The offspring generation process is divided into two phases. The first phase uses the general genetic operators to generate the offspring, and then in the second phase, the selected non-dominated solutions are utilized by the proposed deep sampling method to produce offspring again. Specifically, the proposed deep sampling method makes use of the selected non-dominated solutions to establish search directions at first, then solutions are sampled on them. It is expected to take advantage of both offspring generation schemes, thereby balancing the diversity and convergence of the population. Existing large-scale evolutionary algorithms can easily be extended to our proposed bi-level architecture. The experimental results demonstrate the significant advantages of the proposed architecture and sampling method, in comparison with several state-of-the-art large-scale multi∖many-objective optimization problems in solving LSMOPs with up to 5000 decision variables.

Speed vs Accuracy: Effect on Ligand Pose Accuracy of Varying Box Size and Exhaustiveness in AutoDock Vina.

Structure-based virtual high-throughput screening involves docking chemical libraries to targets of interest. A parameter pertinent to the accuracy of the resulting pose is the root mean square deviation (RMSD) from a known crystallographic structure, i. e., the 'docking power'. Here, using a popular algorithm, Autodock Vina, as a model program, we evaluate the effects of varying two common docking parameters: the box size (the size of docking search space) and the exhaustiveness of the global search (the number of independent runs starting from random ligand conformations) on the RMSD from the PDBbind v2017 refined dataset of experimental protein-ligand complexes. Although it is clear that exhaustiveness is an important parameter, there is wide variation in the values used, with variation between 1 and >100. We, therefore, evaluated a combination of cubic boxes of different sizes and five exhaustiveness values (1, 8, 25, 50, 75, 100) within the range of those commonly adopted. The results show that the default exhaustiveness value of 8 performs well overall for most box sizes. In contrast, for all box sizes, but particularly for large boxes, an exhaustiveness value of 1 led to significantly higher median RMSD (mRMSD) values. The docking power was slightly improved with an exhaustiveness of 25, but the mRMSD changes little with values higher than 25. Therefore, although low exhaustiveness is computationally faster, the results are more likely to be far from reality, and, conversely, values >25 led to little improvement at the expense of computational resources. Overall, we recommend users to use at least the default exhaustiveness value of 8 for virtual screening calculations.