Local Search Algorithm Research Articles

Q-learning guided algorithms for bi-criteria minimization of total flow time and makespan in no-wait permutation flowshops

Combining Deep Reinforcement Learning and meta-heuristic techniques represents a new research direction for enhancing the search capabilities of meta-heuristic methods in the context of production scheduling. Q-learning is a prominent reinforcement learning in which its utilization aims to direct the selection of actions, thus preventing the necessity for a random exploration in the iterative process of the metaheuristics. In this study, we provide Q-learning guided algorithms for the Bi-Criteria No-Wait Flowshop Scheduling Problem (NWFSP). The problem is treated as a bi-criteria combinatorial optimization problem where total flow time and makespan are optimized simultaneously. Firstly, a deterministic mixed-integer linear programming (MILP) model is provided. Then, Q-learning guided algorithms are developed: Bi-Criteria Iterated Greedy Algorithm with Q-Learning (BC-IGQL). Bi-Criteria Block Insertion Heuristic Algorithm with Q-Learning (BC-BIHQL). Moreover, the performance of the proposed Q-learning guided algorithms is compared over a collection of Bi-Criteria Genetic Local Search Algorithms (BC-GLS), Bi-Criteria Iterated Greedy Algorithm (BC-IG), Bi-Criteria Iterated Greedy Algorithm with a Local Search (BC-IGALL) and Bi-Criteria Variable Block Insertion Heuristic Algorithm (BC-VBIH). The complete computational experiment, performed on 480 problem instances of Vallada et al. (2015), which is known as the VRF benchmark set, indicates that the BC-BIHQL and the BC-IGQL algorithms outperform the BC-GLS, BC-IG, BC-IGALL, and BC-VBIH algorithms in comparative performance metrics. More specifically, the proposed BC-BIHQL and BC-IGQL algorithms can yield more non-dominated bi-criteria solutions with the most substantial competitiveness than the remaining algorithms. At the same time, both are competitive with each other on the benchmark problems. Moreover, the BC-IGQL algorithm dominates almost 97% and 99% of the solutions reached by the BC-IG, BC-IGALL, and BC-VBIH algorithms in small and large datasets. Similarly, The BC-BIHQL algorithm dominates almost 98% and 99% of the solutions reached by the BC-IG, BC-IGALL, and BC-VBIH algorithms in small and large datasets, respectively. This means that, among all the features that have been compared, the Q-learning-guided algorithms demonstrate the highest level of competitiveness. The outcomes of this study encourage us to discover many more bi-criteria NWFSPs to reveal the trade-off between other conflicting objectives, such as makespan & the number of early jobs, to overcome various industries' problems.

WordBlitz: An Efficient Hard-Label Textual Adversarial Attack Method Jointly Leveraging Adversarial Transferability and Word Importance

Existing textual attacks mostly perturb keywords in sentences to generate adversarial examples by relying on the prediction confidence of victim models. In practice, attackers can only access the prediction label, meaning that the victim model can easily defend against such hard-label attacks by denying access based on the attack’s frequency. In this paper, we propose an efficient hard-label attack approach, called WordBlitz. First, based on the adversarial transferability, we train a substitute model to initialize the attack parameter set, including a candidate pool and two weight tables of keywords and candidate words. Then, adversarial examples are generated and optimized under the guidance of the two weight tables. During optimization, we design a hybrid local search algorithm with word importance to find the globally optimal solution while updating the two weight tables according to the attack results. Finally, the non-adversarial text generated during perturbation optimization is added to the training of the substitute model as data augmentation to improve the adversarial transferability. Experimental results show that WordBlitz surpasses the baseline in terms of better effectiveness, higher efficiency, and lower cost. Its efficiency is especially pronounced in scenarios with broader search spaces, and its attack success rate on a Chinese dataset is higher than on baselines.

Crop Yield Estimation using Improved Salp Swarm Algorithm based Feature Selection

Crop yield estimation is the art of yield prediction before harvest and it is essential for planning and making conclusive agricultural policies. The forecasting of crop yield is essential in optimal nutrient management, crop insurance, crop market planning and harvest management. However, the crop yield estimation is considered as a challenging task because of huge amount of abundant information exists in the crop data. Therefore, an effective feature selection is required to be developed for removing the redundant attributes. In this research, an Improved Salp Swarm Algorithm (ISSA) based feature selection for an effective crop yield estimation. The Opposition Based Learning (OBL) and Local Search Algorithm (LSA) are incorporated in the ISSA’s initialization and exploitation phase for selecting optimum feature subset. The selected features from the ISSA are used to enhance the classification using Modified Long Short Term Memory (MLSTM) classifier. The performance of the ISSA-MLSTM is analyzed using accuracy, precision, recall, F-score, Nash-Sutcliffe Efficiency Coefficient (NSEC), Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The existing researches such as Ensemble approach and MLSTM are used to evaluate the ISSA-MLSTM. The accuracy of the ISSA-MLSTM is 99% that is high when compared to the MLSTM.

A frequency and two-hop configuration checking-driven local search algorithm for the minimum weakly connected dominating set problem

A frequency and two-hop configuration checking-driven local search algorithm for the minimum weakly connected dominating set problem

Optimizing with Attractor: A Tutorial

This tutorial presents a novel search system—the Attractor-Based Search System (ABSS)—that can solve the Traveling Salesman Problem very efficiently with optimality guarantee. From the perspective of dynamical systems, a heuristic local search algorithm for an NP-complete combinatorial problem is a discrete dynamical system. In a local search system, an attractor drives the search trajectories into the vicinity of a globally optimal point in the solution space, and the convergence of local search trajectories makes the search system become a global and deterministic system. The attractor contains a small set of the most promising solutions to the problem. The attractor can reduce the problem size exponentially, and thus make the exhaustive search feasible. Therefore, this new search paradigm is called optimizing with attractor. The ABSS consists of two search phases: local search phase and exhaustive search phase. The local search process is used to quickly construct the attractor in the solution space, and the exhaustive search process is used to completely search the attractor to identify the optimal solution. Therefore, the exact optimal solution can be found quickly by combining local search and exhaustive search. This tutorial introduces the concept of an attractor in a local search system, and describes the process of optimizing with the attractor, using the Traveling Salesman Problem as the study platform.

Mixed-model sequencing with reinsertion of failed vehicles: a case study for automobile industry

In the automotive industry, some vehicles, failed vehicles, cannot be produced according to the planned schedule due to some reasons such as material shortage, paint failure, etc. These vehicles are pulled out of the sequence, potentially resulting in an increased work overload. On the other hand, the reinsertion of failed vehicles is executed dynamically as suitable positions occur. In case such positions do not occur enough, either the vehicles waiting for reinsertion accumulate or reinsertions are made to worse positions by sacrificing production efficiency. This study proposes a bi-objective two-stage stochastic program and formulation improvements for a mixed-model sequencing problem with stochastic product failures and integrated reinsertion process. Moreover, an evolutionary optimisation algorithm, a two-stage local search algorithm, and a hybrid approach are developed. Numerical experiments over a case study show that while the hybrid algorithm better explores the Pareto front representation, the local search algorithm provides more reliable solutions regarding work overload objective. Finally, the results of the dynamic reinsertion simulations show that we can decrease the work overload by 20% while significantly decreasing the waiting time of the failed vehicles by considering vehicle failures and integrating the reinsertion process into the mixed-model sequencing problem.

Parameter-less Pareto local search for multi-objective neural architecture search with the Interleaved Multi-start Scheme

With the emerging deployment of deep neural networks, such as in mobile devices and autonomous cars, there is a growing demand for neural architecture search (NAS) to automatically design powerful network architectures. It is more reasonable to formulate NAS as a multi-objective optimization problem. In addition to prediction performance, multi-objective NAS (MONAS) problems take into account other criteria like the number of parameters and inference latency. Multi-objective evolutionary algorithms (MOEAs) are the preferred approach for tackling MONAS due to their effectiveness in dealing with multi-objective optimization problems. Recently, local search-based NAS algorithms have demonstrated their efficiency over MOEAs for MONAS problems. However, their performance has been only verified on bi-objective NAS problems. In this article, we propose a local search algorithm for multi-objective NAS (LOMONAS), an efficient local search framework for solving not only bi-objective NAS problems but also NAS problems having more than two objectives. We additionally present a parameter-less version of LOMONAS, namely IMS-LOMONAS, by combining LOMONAS with the Interleaved Multi-start Scheme (IMS) to help NAS practitioners avoid manual control parameter settings. Experimental results from a series of benchmark problems in the CEC’23 Competition demonstrate the competitiveness of LOMONAS and IMS-LOMONAS compared to MOEAs in tackling MONAS within both small-scale and large-scale search spaces. Source code is available at: https://github.com/ELO-Lab/IMS-LOMONAS.

Joint optimization of inventory replenishment and rationing policies for an omnichannel store with both in-store and online demands

Boosted by the COVID-19 pandemic, omnichannel retail has become a popular retail model and is expected to further grow. We study inventory replenishment and rationing of a product in an omnichannel store that fulfils both in-store and online demands. Both demands follow Poisson process. As each in-store demand must be satisfied immediately whereas each online order has a delivery time window, the two demand classes have different behaviors facing stockout: offline demand is lost while online order is backlogged with backorder cost charged only if it is delivered late. The inventory replenishment and rationing between the two demand classes in this omnichannel inventory system are controlled by a continuous review (Q, r) policy and a critical level policy, respectively. The intertwined nature of inventory replenishment and rationing in this system makes its analytical cost evaluation and its joint optimization of the two policies challenging. After deriving an analytical formula for evaluating the expected cost of the system, two heuristic algorithms are proposed to jointly optimize the two policies. Numerical experiments on randomly generated instances demonstrate that the multi-start local search algorithm and the scatter search algorithm can obtain a near-optimal solution quickly with percentage cost gap no larger than 0.81% and 0.18%, respectively, and the critical level policy can yield cost savings up to 10.91% compared to the first-come, first-served policy. Finally, a sensitivity analysis is conducted to evaluate the influences of system parameters on the optimal critical level.

Application of Neural Network in Seismic Risk Survey

The accuracy of inversion data is determined by the reliability of linear inversion method. Therefore, the study of nonlinear seismic inversion method is a meaningful work. This paper mainly aims at the application of nonlinear neural network inversion method in Fujian Risk Survey, the following work has been accomplished: (1) The theory and algorithm of the conventional multilayer feedforward neural network and some problems in the design of the model are analyzed and summarized. Aiming at the gradient descent function search algorithm in the BP algorithm, a method for global optimization is proposed by combining simulated annealing (global search algorithm) with Conjugate gradient method (local search algorithm) , the principle and implementation of simulated annealing and conjugate gradient algorithm are analyzed in detail. (2) Based on the analysis of conventional multilayer feedforward neural networks, the model structure and basic principle of probabilistic neural networks are studied. The object function search algorithm and its realization are analyzed. The method is applied to the risk elimination prediction for the first time, and the results are satisfactory. (3) The method and principle of inversion of reservoir parameters by using various seismic attribute information and neural network technology are discussed. In order to make full use of the effective information contained in seismic signal, the method of seismic attribute extraction and optimization is studied, the principle and implementation method of attribute optimal ranking by stepwise recursive method and selecting the most important attribute combination by interactive verification method are presented.

Goal-conflict identification based on local search and fast boundary-condition verification based on incremental satisfiability filter

Identifying boundary conditions (BCs) is of fundamental importance for goal-conflict analysis. BCs are able to capture particular combinations of circumstances that make some special conflicts, namely goal divergences, in which the goals of the requirement cannot be satisfied as a whole. Unfortunately, existing approaches only handle small-scale cases within a reasonable running time and do not consider how to search more general BCs because of the search space explosion and the high computational cost of verifying BCs. In this paper, we observed that the BCs identified by the state-of-the-art (SOTA) approach share similarities in structure. It inspires us to develop a BC identification approach, named LOGION, based on local search, to better exploit the similarity in structures. Specifically, we utilize the structural similarity of BCs to construct the neighborhood relation of our local search algorithm. Besides, we design strengthening and weakening operators of formulae to capture the semantics of more general BCs in local searching. In order to boost the BC verification, we propose a lasso-based incremental satisfiability filter (LISF). The primary intuition behind LISF is using trace checking as a light-weight pre-checking, and if it is not successful, to call the satisfiability checking. We conduct extensive experiments to evaluate LOGION, comparing against SOTA approaches. Experimental results show that LOGION produces about 1 order of magnitude more general BCs than the SOTA approaches. It is also shown that LISF significantly boosts the speed, especially for large-scale cases (up to about 1 order of magnitude faster on average).

Research on the performance of harmony search with local search algorithms for solving flexible job-shop scheduling problem

Combining the harmony search algorithm (HS) with the local search algorithm (LS) can prevent the HS from falling into a local optimum. However, how LS affects the performance of HS has not yet been studied systematically. Therefore, in this paper, it is first proposed to combine four frequently used LS with HS to obtain several search algorithms (HSLSs). Then, by taking the flexible job-shop scheduling problem (FJSP) as an example and considering decoding times, study how the parameters of HSLSs affect their performance, where the performance is evaluated by the difference rate based on the decoding times. The simulation results mainly show that (I) as the harmony memory size (HMS) gradually increases, the performance of HSLSs first increases rapidly and then tends to remain unchanged, and HMS is not the larger the better; (II) as harmony memory considering rate increases, the performance continues to improve, while the performance of pitch adjusting rate on HSLSs goes to the opposite; Finally, more benchmark instances are also used to verify the effectiveness of the proposed algorithms. The results of this paper have a certain guiding significance on how to choose LS and other parameters to improve HS for solving FJSP.

Team orienteering with possible multiple visits: Mathematical model and solution algorithms

This study addresses a new team orienteering problem in which each point can be visited possibly multiple times with decreasing scores. The problem is to determine a set of routes from a starting to a finishing point within an allowed travel time limit for the objective of maximizing the total collected score. A mixed integer programming model is developed to represent the problem mathematically. Then, due to the complexity of the problem, a basic iterated local search (ILS) algorithm is proposed that incorporates the ordinary perturbation and local search operators. Then, it is improved in two ways: (a) an extended ILS algorithm with multi-visit based local search operators and (b) a hybrid ILS algorithm with a simulated annealing technique to permit non-improving moves in the local search step. To test the performance of the ILS algorithms, computational experiments were done on modified benchmark instances under the cases of linear, convex and concave score decreases, and the results show that the hybrid ILS algorithm improves the others significantly and also gives near optimal solutions for small sized test instances. Finally, a case study was done for a naval warship reconnaissance operation, and the results are reported.

Test Center Location Problem: A Bi-Objective Model and Algorithms

The optimal placement of healthcare facilities, including the placement of diagnostic test centers, plays a pivotal role in ensuring efficient and equitable access to healthcare services. However, the emergence of unique complexities in the context of a pandemic, exemplified by the COVID-19 crisis, has necessitated the development of customized solutions. This paper introduces a bi-objective integer linear programming model designed to achieve two key objectives: minimizing average travel time for individuals visiting testing centers and maximizing an equitable workload distribution among testing centers. This problem is NP-hard and we propose a customized local search algorithm based on the Voronoi diagram. Additionally, we employ an ϵ-constraint approach, which leverages the Gurobi solver. We rigorously examine the effectiveness of the model and the algorithms through numerical experiments and demonstrate their capability to identify Pareto-optimal solutions. We show that while the Gurobi performs efficiently in small-size instances, our proposed algorithm outperforms it in large-size instances of the problem.

Enhance Diversified Top-k MaxSAT Solving by Incorporating New Strategy for Generating Diversified Initial Assignments (Student Abstract)

The Diversified Top-k MaxSAT (DTKMS) problem is an extension of MaxSAT. The objective of DTKMS is to find k feasible assignments of a given formula, such that each assignment satisfies all hard clauses and the k assignments together satisfy the maximum number of soft clauses. This paper presents a local search algorithm, DTKMS-DIA, which incorporates a new approach to generating initial assignments. Experimental results indicate that DTKMS-DIA can achieve attractive performance on 826 instances compared with state-of-the-art solvers.

Optimization of vehicle charging and dynamic relocation in free-floating electric carsharing systems with advanced reservations

The electric free-floating carsharing (FFCS) system is an emerging type of shared mobility and is widely adopted in urban areas due to its eco-friendliness, effectiveness, and flexibility. However, as most trips in FFCS are uni-directional, the distributions of vehicles in FFCS are usually spatially uneven. The operators of FFCS must relocate cars during the operation period to maintain the availability of cars to the users and improve the profitability of the system for dealing with imbalanced vehicle distribution caused by uneven demand. To optimize the relocation operations for FFCS operators, this study formulates the problem as a mixed-integer linear programming (MILP) model, which simultaneously addresses dynamic relocation, battery charging of electric vehicles, and scheduling of relocation staff in an FFCS with advanced reservation. A novel, efficient algorithm using a new structure of relocation tuples (RTs) is proposed to solve the problem that facilitates the search for feasible solutions using an RT exchange-based local search algorithm. Acceleration techniques are introduced to strengthen the solution search in large-scale scenarios. Numerical examples are used to evaluate the performance of the solution algorithm and validate the benefit of the proposed methodology. This shows that the methodology could be used by operators for decision-making at different levels.

Sensitive remote homology search by local alignment of small positional embeddings from protein language models

Accurately detecting distant evolutionary relationships between proteins remains an ongoing challenge in bioinformatics. Search methods based on primary sequence struggle to accurately detect homology between sequences with less than 20% amino acid identity. Profile- and structure-based strategies extend sensitive search capabilities into this twilight zone of sequence similarity but require slow pre-processing steps. Recently, whole-protein and positional embeddings from deep neural networks have shown promise for providing sensitive sequence comparison and annotation at long evolutionary distances. Embeddings are generally faster to compute than profiles and predicted structures but still suffer several drawbacks related to the ability of whole-protein embeddings to discriminate domain-level homology, and the database size and search speed of methods using positional embeddings. In this work, we show that low-dimensionality positional embeddings can be used directly in speed-optimized local search algorithms. As a proof of concept, we use the ESM2 3B model to convert primary sequences directly into the 3D interaction (3Di) alphabet or amino acid profiles and use these embeddings as input to the highly optimized Foldseek, HMMER3, and HH-suite search algorithms. Our results suggest that positional embeddings as small as a single byte can provide sufficient information for dramatically improved sensitivity over amino acid sequence searches without sacrificing search speed.

Sensitive remote homology search by local alignment of small positional embeddings from protein language models.

Accurately detecting distant evolutionary relationships between proteins remains an ongoing challenge in bioinformatics. Search methods based on primary sequence struggle to accurately detect homology between sequences with less than 20% amino acid identity. Profile- and structure-based strategies extend sensitive search capabilities into this twilight zone of sequence similarity but require slow pre-processing steps. Recently, whole-protein and positional embeddings from deep neural networks have shown promise for providing sensitive sequence comparison and annotation at long evolutionary distances. Embeddings are generally faster to compute than profiles and predicted structures but still suffer several drawbacks related to the ability of whole-protein embeddings to discriminate domain-level homology, and the database size and search speed of methods using positional embeddings. In this work, we show that low-dimensionality positional embeddings can be used directly in speed-optimized local search algorithms. As a proof of concept, we use the ESM2 3B model to convert primary sequences directly into the 3D interaction (3Di) alphabet or amino acid profiles and use these embeddings as input to the highly optimized Foldseek, HMMER3, and HH-suite search algorithms. Our results suggest that positional embeddings as small as a single byte can provide sufficient information for dramatically improved sensitivity over amino acid sequence searches without sacrificing search speed.

An improved particle swarm optimization algorithm for scheduling tasks in cloud environment

AbstractCloud computing provide services dynamically according to the contract between service providers and users. However, Inappropriateness of scheduling task on VMs can lead huge resource waste and load unbalance, which becomes a seriously challenging problem. Current Swarm intelligence algorithms like genetic algorithm (GA), particle swarm optimization (PSO) are combination of random initialization and local search algorithm. It avoids inconsistent results for different problem instances. However, existing Swarm intelligence works sometimes search the optima without analysing task scheduling situations comprehensively, global search efficiency is low and convergence is too early. In this paper, we propose SNSK‐IPSO algorithm, which develops as a two‐phases algorithm: enumerating all distributed solutions between VMs and tasks, finding the optimal solution through IPSO. It not only minimizes the execution time, but also improves resource utilization and load balance. Several experiments demonstrate that our novel algorithm outperforms others in terms of achieving load balance, higher resource utilization and lower execution times.

Iterated local search with ejection chains for the space-free multi-row facility layout problem

This paper presents an iterated local search algorithm based on ejection chains to solve the space-free multi-row facility layout problem. The aim of this problem is to find a non-overlapping layout of facilities on a given number of rows such that there is no space between two adjacent facilities. In addition, the left-most facility of the arrangement must have zero abscissa. Our algorithm looks for a local optimal solution by continuously alternating ejection moves and trial moves to form an ejection chain. Since the ejection chain can compound simple moves to create more complex and powerful moves, it has a greater chance to discover better solutions. Additionally, we propose a directional perturbation strategy to produce a solution of both high quality and good diversity. The idea is to calculate the score of each facility based on the location concentration of facilities in the elite solution set and the historical perturbation times of facilities in the search process, and to select the facility with the highest score for perturbation. Experimental results on four benchmark sets containing a total of 290 instances reveal that the proposed algorithm performs better for 131 (45.2%) instances in terms of best solution values and for 196 (80.0%) instances in terms of average solution values than a recently reported state-of-the-art algorithm.

Theoretical Exploration of Flexible Transmitter Model.

Neural network models generally involve two important components, i.e., network architecture and neuron model. Although there are abundant studies about network architectures, only a few neuron models have been developed, such as the MP neuron model developed in 1943 and the spiking neuron model developed in the 1950s. Recently, a new bio-plausible neuron model, flexible transmitter (FT) model (Zhang and Zhou, 2021), has been proposed. It exhibits promising behaviors, particularly on temporal-spatial signals, even when simply embedded into the common feedforward network architecture. This article attempts to understand the properties of the FT network (FTNet) theoretically. Under mild assumptions, we show that: 1) FTNet is a universal approximator; 2) the approximation complexity of FTNet can be exponentially smaller than those of commonly used real-valued neural networks with feedforward/recurrent architectures and is of the same order in the worst case; and 3) any local minimum of FTNet is the global minimum, implying that it is possible to identify global minima by local search algorithms.