Low-level Heuristics Research Articles

An Estimation of Distribution Algorithm-Based Hyper-Heuristic for the Distributed Assembly Mixed No-Idle Permutation Flowshop Scheduling Problem

The distributed assembly mixed no-idle permutation flowshop scheduling problem (DAMNIPFSP), a common occurrence in modern industries like integrated circuit production, ceramic frit production, fiberglass processing, and steel-making, is a new model that considers mixed machines with no-idle restrictions as well as conventional machines. This article introduces an estimation of distribution algorithm-based hyper-heuristic (EDA-HH) to solve the DAMNIPFSP. Ten simple heuristic rules as low-level operations are utilized to search the solution space. The estimation of distribution algorithm is integrated into the framework of hyper-heuristic as the high-level strategy to control the low-level heuristics sequence in the solution space. The destruction and construction procedures are conducted on products and jobs in order to enhance the exploitation competence of EDA-HH. The computational simulation is carried out and the experimental results show that the proposed EDA-HH is significantly superior to the competitors in the statistical sense. The results of the 810 large-scale problem instances show the effectiveness of the EDA-HH in solving the DAMNIPFSP. Moreover, the CPLEX solver is utilized to verify the correctness of the model with some small instances.

Q-learning-based hyper-heuristic evolutionary algorithm for the distributed assembly blocking flowshop scheduling problem

Distributed shop scheduling problems (DSSPs) have attracted increasing interest in recent years due to the technical trends of smart manufacturing and Industry 4.0. The distributed assembly blocking flowshop scheduling problem (DABFSP) is a critical class of DSSPs with widespread applications in modern supply chains and manufacturing systems. In this paper, a Q-learning-based hyper-heuristic evolutionary algorithm (QLHHEA) is proposed to solve DABFSP with the objective of minimizing the makespan. Firstly, a mathematical model of DABFSP is formulated, and two insertion-based speedup strategies are devised to conserve the computational cost of evaluating solutions and to accelerate the search efficiency. Secondly, a problem-specific constructive heuristic is developed to produce high-quality initial solutions. Thirdly, twelve efficient heuristics are designed to construct low-level heuristics (LLHs). The Q-learning-based evolutionary algorithm is applied as a high-level strategy to manipulate the LLHs, which are then executed in order to search the solution space. Moreover, suitable solution encoding and decoding schemes are provided to produce feasible scheduling schedules. The design of experiments is implemented to investigate the impact of the parameters. Finally, a comprehensive comparison campaign is carried out based on a total of 1710 well-known instances to evaluate the efficacy of the proposed algorithm against several state-of-the-art algorithms. Experimental results and statistical analysis show that QLHHEA significantly outperforms the existing algorithms by a significant margin, demonstrating the effectiveness and efficiency of QLHHEA in solving DABFSP.

A deep reinforcement learning based hyper-heuristic for modular production control

In nowadays production, fluctuations in demand, shortening product life-cycles, and highly configurable products require an adaptive and robust control approach to maintain competitiveness. This approach must not only optimise desired production objectives but also cope with unforeseen machine failures, rush orders, and changes in short-term demand. Previous control approaches were often implemented using a single operations layer and a standalone deep learning approach, which may not adequately address the complex organisational demands of modern manufacturing systems. To address this challenge, we propose a hyper-heuristics control model within a semi-heterarchical production system, in which multiple manufacturing and distribution agents are spread across pre-defined modules. The agents employ a deep reinforcement learning algorithm to learn a policy for selecting low-level heuristics in a situation-specific manner, thereby leveraging system performance and adaptability. We tested our approach in simulation and transferred it to a hybrid production environment. By that, we were able to demonstrate its multi-objective optimisation capabilities compared to conventional approaches in terms of mean throughput time, tardiness, and processing of prioritised orders in a multi-layered production system. The modular design is promising in reducing the overall system complexity and facilitates a quick and seamless integration into other scenarios.

Guided Intelligent Hyper-Heuristic Algorithm for Critical Software Application Testing Satisfying Multiple Coverage Criteria

This paper proposes a novel algorithm that combines symbolic execution and data flow testing to generate test cases satisfying multiple coverage criteria of critical software applications. The coverage criteria considered are data flow coverage as the primary criterion, software safety requirements, and equivalence partitioning as sub-criteria. black The characteristics of the subjects used for the study include high-precision floating-point computation and iterative programs. The work proposes an algorithm that aids the tester in automated test data generation, satisfying multiple coverage criteria for critical software. The algorithm adapts itself and selects different heuristics based on program characteristics. The algorithm has an intelligent agent as its decision support system to accomplish this adaptability. Intelligent agent uses the knowledge base to select different low-level heuristics based on the current state of the problem instance during each generation of genetic algorithm execution. The knowledge base mimics the expert’s decision in choosing the appropriate heuristics. black The algorithm outperforms by accomplishing 100% data flow coverage for all subjects. In contrast, the simple genetic algorithm, random testing and a hyper-heuristic algorithm could accomplish a maximum of 83%, 67% and 76.7%, respectively, for the subject program with high complexity. black The proposed algorithm covers other criteria, namely equivalence partition coverage and software safety requirements, with fewer iterations. black The results reveal that test cases generated by the proposed algorithm are also effective in fault detection, with 87.2% of mutants killed when compared to a maximum of 76.4% of mutants killed for the complex subject with test cases of other methods.

Solving the Multi-robot task allocation with functional tasks based on a hyper-heuristic algorithm

Multi-robot task allocation (MRTA) problem is a classical problem in multi-robot systems. The most common assumption in MRTA is that all the tasks need to be completed with the time cost as small as possible. It is worth noting that some tasks may be optional in some real-world situations, i.e., the robots do not necessarily need to complete all tasks. These optional tasks do not limit the achievement of the goal, but the completion of these tasks will lead to some functional effects (e.g., making other tasks easier to be completed). Therefore, these optional tasks can be called “functional task”. Different from functional tasks, if a task must be completed, this task can be called “compulsory task”. In this paper, we study the problem where the robots need to minimize the time cost of completing all compulsory tasks and can selectively complete some functional tasks. The existence of the functional tasks greatly increases the solution space of MRTA, because the functional tasks should be firstly suitably selected and then suitably allocated. Existing related algorithms are usually based on the assumption that all tasks must be allocated. Thus, these algorithms cannot suitably deal with functional tasks. We analyze the characteristics of this problem and present a new hyper-heuristic algorithm. The low-level heuristic (LLH) in hyper-heuristic is designed to score the functional tasks by using influence diffusion model. The high-level strategy (HLS) seeks the optimal values of the key parameters in the influence diffusion model based on particle swarm optimization (PSO). Extensive simulated experiments are presented to comprehensively analyze the proposed algorithm. The proposed hyper-heuristic is compared with greedy algorithm and two meta-heuristic algorithms. Based on the simulated data, it is known that the hyper-heuristic algorithm can outperform the benchmark algorithms especially when the number of functional tasks is large.

Large-State Reinforcement Learning for Hyper-Heuristics

Hyper-heuristics are a domain-independent problem solving approach where the main task is to select effective chains of problem-specific low-level heuristics on the fly for an unseen instance. This task can be seen as a reinforcement learning problem, however, the information available to the hyper-heuristic is very limited, usually leading to very limited state representations. In this work, for the first time we use the trajectory of solution changes for a larger set of features for reinforcement learning in the novel hyper-heuristic LAST-RL (Large-State Reinforcement Learning). Further, we introduce a probability distribution for the exploration case in our epsilon-greedy policy that is based on the idea of Iterated Local Search to increase the chance to sample good chains of low-level heuristics. The benefit of the collaboration of our novel components is shown on the academic benchmark of the Cross Domain Heuristic Challenge 2011 consisting of six different problem domains. Our approach can provide state-of-the-art results on this benchmark where it outperforms recent hyper-heuristics based on reinforcement learning, and also demonstrates high performance on a benchmark of complex real-life personnel scheduling domains.

Multi-armed bandit-based hyper-heuristics for combinatorial optimization problems

There are significant research opportunities in the integration of Machine Learning (ML) methods and Combinatorial Optimization Problems (COPs). In this work, we focus on metaheuristics to solve COPs that have an important learning component. These algorithms must explore a solution space and learn from the information they obtain in order to find high-quality solutions. Among the metaheuristics, we study Hyper-Heuristics (HHs), algorithms that, given a number of low-level heuristics, iteratively select and apply heuristics to a solution. The HH we consider has a Markov model to produce sequences of low-level heuristics, which we combine with a Multi-Armed Bandit Problem (MAB)-based method to learn its parameters. This work proposes several improvements to the HH metaheuristic that yields a better learning for solving problem instances. Specifically, this is the first work in HHs to present Exponential Weights for Exploration and Exploitation (EXP3) as a learning method, an algorithm that is able to deal with adversarial settings. We also present a case study for the Vehicle Routing Problem with Time Windows (VRPTW), for which we include a list of low-level heuristics that have been proposed in the literature. We show that our algorithms can handle a large and diverse list of heuristics, illustrating that they can be easily configured to solve COPs of different nature. The computational results indicate that our algorithms are competitive methods for the VRPTW (2.16% gap on average with respect to the best known solutions), demonstrating the potential of these algorithms to solve COPs. Finally, we show how algorithms can even detect low-level heuristics that do not contribute to finding better solutions to the problem.

Big Data Cyber Security Using Machine Learning

It is well-known that cyber protection in the context of big data is a crucial issue and also presents a significant challenge to the research study field. The use of artificial intelligence algorithms has been suggested as a potential solution for complex information security issues. Support vector machines (SVMs), one of these techniques, have shown outstanding performance on several category problems. However, in order to construct an effective SVM, the user must first determine the correct SVM configuration, which is a difficult task that necessitates expert knowledge as well as a substantial amount of hands-on initiative for trial and error. In this study, precision and design intricacy are treated as two conflicting goals in our formulation of the SVM arrangement technique as a bi-objective optimisation problem. We propose a brand-new, issue-domain-independent hyper-heuristic framework for bi-objective optimisation. This is the first time a hyper-heuristic for this issue has been developed. The suggested hyper-heuristic framework consists of both high-level and low-level heuristics. The high-level strategy makes decisions on which low-level heuristics to employ in order to create a fresh SVM setup based on the search performance. Each of the low-level heuristics successfully locates the SVM layout search region using a distinct set of rules. The suggested structure adaptively integrates Pareto-based and disintegration-based strategy toughness to approximate the Pareto collection of SVM configurations in order to handle bi-objective optimisation. The effectiveness of the suggested structure has been evaluated in relation to two cyber security concerns: the detection of anomalous intrusions and Microsoft malware's large information category. The obtained results show that the suggested framework is quite effective, if not superior, when compared to its competitors and other formulas.

A Hyperheuristic With Q-Learning for the Multiobjective Energy-Efficient Distributed Blocking Flow Shop Scheduling Problem.

Carbon peaking and carbon neutrality, which are the significant national strategy for sustainable development, have attracted considerable attention from production enterprises. In this study, the energy consumption is considered in the distributed blocking flow shop scheduling problem (DBFSP). A hyperheuristic with Q -learning (HHQL) is presented to address the energy-efficient DBFSP (EEDBFSP). Q -learning is employed to select an appropriate low-level heuristic (LLH) from a predesigned LLH set according to historical information fed back by LLH. An initialization method, which considers both total tardiness (TTD) and total energy consumption (TEC), is proposed to construct the initial population. The ε -greedy strategy is introduced to utilize the learned knowledge while retaining a certain degree of exploration in the process of selecting LLH. The acceleration operation of the job on the critical path is designed to optimize TTD. The deceleration operation of the job on the noncritical path is designed to optimize TEC. The statistical and computational experimentation in an extensive benchmark testified that the HHQL outperforms the other comparison algorithm regarding efficiency and significance in solving EEDBFSP.

Sequencing single machine multiple-class customer order jobs using heuristics and improved simulated annealing algorithms

The multiple job class scheduling problem arises in contexts where a group of jobs belong to multiple classes and in which if all jobs in the same class are operated together, extra setup times would not be needed. On the other hand, the customer order scheduling problem focuses on finishing all jobs from the same order at the same time in order to reduce shipping costs. However, works on customer orders coupled with class setup times do not appear often in the literature. Hence we address here a bicriteria single machine customer order scheduling problem together with multiple job classes. The optimality criterion minimizes a linear combination of the sum of the ranges and sum of tardiness of all customer orders. In light of the high complexity of the concerned problem, we propose a lower bound formula and a property to be used in a branch-and-bound method for optimal solutions. To find approximate solutions, we then propose four heuristics together with a local search method, four cloudy theoretical simulated annealing and a cloudy theoretical simulated annealing hyperheuristic along with five low-level heuristics. The simulation results of the proposed heuristics and algorithms are analyzed.

Q-Learning-Based Hyperheuristic Evolutionary Algorithm for Dynamic Task Allocation of Crowdsensing.

Task allocation is a crucial issue of mobile crowdsensing. The existing crowdsensing systems normally select the optimal participants giving no consideration to the sudden departure of mobile users, which significantly affects the sensing quality of tasks with a long sensing period. Furthermore, the ability of a mobile user to collect high-precision data is commonly treated as the same for different types of tasks, causing the unqualified data for some tasks provided by a competitive user. To address the issue, a dynamic task allocation model of crowdsensing is constructed by considering mobile user availability and tasks changing over time. Moreover, a novel indicator for comprehensively evaluating the sensing ability of mobile users collecting high-quality data for different types of tasks at the target area is proposed. A new Q -learning-based hyperheuristic evolutionary algorithm is suggested to deal with the problem in a self-learning way. Specifically, a memory-based initialization strategy is developed to seed a promising population by reusing participants who are capable of completing a particular task with high quality in the historical optima. In addition, taking both sensing ability and cost of a mobile user into account, a novel comprehensive strength-based neighborhood search is introduced as a low-level heuristic (LLH) to select a substitute for a costly participant. Finally, based on a new definition of the state, a Q -learning-based high-level strategy is designed to find a suitable LLH for each state. Empirical results of 30 static and 20 dynamic experiments expose that this hyperheuristic achieves superior performance compared to other state-of-the-art algorithms.

Surgical cases assignment problem using an efficient genetic programming hyper-heuristic

The surgical case assignment problem (SCAP) is vital to the operating room planning problem. Although several methods have been applied, the solution accuracy can be improved further. In this paper, an efficient genetic programming hyper-heuristic (GP-HH) algorithm is proposed for the SCAP to minimize the total operating cost. First, eight simple and adaptive heuristic rules are devised to constitute a set of low-level heuristics (LLHs). Second, genetic programming is employed as a high-level heuristic to dynamically manage LLHs applied to the solution domain. Third, effective solution encoding and the corresponding decoding schemes are developed to represent individuals and construct valid schedules. To investigate the influence of parameter settings, we performed a design-of-experiment (DOE). The effectiveness of GP-HH is executed on a typical benchmark dataset. The experimental results demonstrate the superiority of the proposed GP-HH scheme over existing approaches.

Study on location-allocation problem and algorithm for emergency supplies considering timeliness and fairness

To address the situation of insufficient reserve of resources in the initial stage of emergency rescue, a multi-objective location-allocation model for emergency supplies with timeliness and fairness being considered concurrently is proposed in this paper. The first objective is defined as time cost, including transport time cost and waiting time cost. The second objective is defined as the number of short supplies. The allocation of supplies is taken into account together with the urgency degree in a disaster area (DA), and the DAs with different urgency degrees are given different minimum allocation quantities of supplies to achieve fairness to a maximum extent. To solve this complex problem, a multi-objective hyper-heuristic (MOHH) optimization framework based on an evolutionary algorithm is proposed in this paper. In the framework, twelve low-level heuristics (LLHs) are designed with the actual information in the problem field being taken into account, and an online learning-based choice strategy is designed to choose high-quality and efficient LLHs. In addition, three acceptance criteria (AC) based on the D matrix are put forward to improve the performance of the MOHH framework. It is verified through comparative analysis experiments that the LLHs and the model are effective and the performance of the proposed multi-objective algorithm is better than that of NSGA-III and MOPSO.

LS-HH: A Learning-Based Selection Hyper-Heuristic for Distributed Heterogeneous Hybrid Blocking Flow-Shop Scheduling

As the development of economic globalization, the distributed manufacturing has become common in modern industries. The scheduling of production resources in multiple production centers becomes an emerging topic. This paper is the first attempt to address a distributed heterogeneous hybrid blocking flow-shop scheduling problem (DHHFSP-B) with the minimization of makespan. Compared with the traditional single flow-shop scheduling, DHHFSP-B considers the collaborative production of multiple hybrid flow lines with heterogeneous layout and processing performance as well as no intermediate buffers. We firstly present a mixed-integer linear programming model to formulate DHHFSP-B and then propose a learning-based selection hyper-heuristic framework (LS-HH) for solving it. The LS-HH contains high-level strategy and low-level heuristics. In the high-level strategy, a learning probability model is built to provide the guidance to choose the suitable perturbation heuristic during the optimization process. A simulated annealing-like move acceptance is employed to determine the updating of incumbent domain solution and prevent the search from trapping into local optimum. In the low-level heuristics, a constructive heuristic is proposed based on a novel assignment rule to create the initial domain solution. Four problem-specific perturbation heuristics and a variable neighborhood search-based improvement operator are employed to search the solution space. A comprehensive computational experiment is conducted. The comparative results show that the LS-HH significantly outperforms the Gurobi solver and several closely relevant optimization methods in solving the DHHFSP-B.

A general deep reinforcement learning hyperheuristic framework for solving combinatorial optimization problems

Many problem-specific heuristic frameworks have been developed to solve combinatorial optimization problems, but these frameworks do not generalize well to other problem domains. Metaheuristic frameworks aim to be more generalizable compared to traditional heuristics, however their performances suffer from poor selection of low-level heuristics (operators) during the search process. An example of heuristic selection in a metaheuristic framework is the adaptive layer of the popular framework of Adaptive Large Neighborhood Search (ALNS). Here, we propose a selection hyperheuristic framework that uses Deep Reinforcement Learning (Deep RL) as an alternative to the adaptive layer of ALNS. Unlike the adaptive layer which only considers heuristics’ past performance for future selection, a Deep RL agent is able to take into account additional information from the search process, e.g., the difference in objective value between iterations, to make better decisions. This is due to the representation power of Deep Learning methods and the decision making capability of the Deep RL agent which can learn to adapt to different problems and instance characteristics. In this paper, by integrating the Deep RL agent into the ALNS framework, we introduce Deep Reinforcement Learning Hyperheuristic (DRLH), a general framework for solving a wide variety of combinatorial optimization problems and show that our framework is better at selecting low-level heuristics at each step of the search process compared to ALNS and a Uniform Random Selection (URS). Our experiments also show that while ALNS can not properly handle a large pool of heuristics, DRLH is not negatively affected by increasing the number of heuristics.

An effective hyper heuristic-based memetic algorithm for the distributed assembly permutation flow-shop scheduling problem

In this paper, an effective Hyper Heuristic-based Memetic Algorithm (HHMA) is proposed to solve the Distributed Assembly Permutation Flow-shop Scheduling Problem (DAPFSP) with the objective of minimizing the maximum completion time. A novel searching-stage-based solution representation scheme is presented for both improving the search efficiency and maintaining potential solutions. In the global search stage, Estimation of Distribution Algorithm (EDA) is employed as the high level strategy of EDA-based Hyper Heuristic (EDAHH) to find promising product sequences for further exploitation. Based on the newly found knowledge of critical-products, several efficient Low-Level Heuristics (LLHs) are well designed to construct the LLH set so that the powerful exploration ability of the EDAHH can be guaranteed. A simulated-annealing-like type of acceptance criterion is also embedded into each LLH to avoid premature convergence. Then a Critical-Products-based Referenced Local Search (CP-RLS) method is proposed to improve the quality of superior sub-population by operating on the sub-job-sequences derived from the critical products. The benefit of the presented CP-RLS lies in the excellent exploitation ability with substantially reduced computational cost. Finally, performance evaluation and comparison are both carried out on a benchmark set and the results demonstrate the superiority of HHMA over the state-of-the-art algorithms for the DAPFSP.

A generality analysis of multiobjective hyper-heuristics

Selection hyper-heuristics have emerged as high level general-purpose search methodologies that mix and control a set of low-level (meta) heuristics. Previous empirical studies over a range of single objective optimisation problems have shown that the number and type of low-level (meta) heuristics used are influential to the performance of selection hyper-heuristics. In addition, move acceptance strategies play an important role and can significantly affect the overall performance of a hyper-heuristic. In this paper, we introduce an adapted variant of an existing learning automata based multiobjective hyper-heuristic from the literature. We investigate the performance and generality level of the proposed method, and another learning automata based selection hyper-heuristic, operating over a search space of multiobjective evolutionary algorithms (MOEAs) across two well-known multiobjective optimisation benchmarks. The experimental results demonstrate that, regardless of the number and type of low-level metaheuristics available, the learning automata based hyper-heuristics outperform each constituent MOEA individually, and an online learning and random choice selection hyper-heuristic from the literature. This performance and generality is shown to be consistent across a number of different move acceptance strategies.

Effects of Heuristic Type, User Interaction Level, and Evaluator’s Characteristics on Usability Metrics of Augmented Reality (AR) User Interfaces

Heuristic evaluation is a famous usability inspection method often used in the assessment of Augmented Reality (AR) systems. Usability heuristics can be characterized as general or specific to a domain. During comparative studies between a new and a traditional set of heuristics, the focus is often around performance metrics (number of problems, time, errors), and little attention is given to the experience itself (i.e., evaluator’s overall satisfaction toward utilization of the set of heuristics). In addition, limited research has been done on the application of heuristics to evaluate interfaces with different levels of interaction, along with how a longer set of low-level heuristics compares against a traditionally-sized set of high-level heuristics. In this study, recruited participants (mostly university students) performed a heuristic evaluation using either a set of 110 specific heuristics (SH) or a set of 23 general heuristics (GH) to evaluate two versions of an AR application, designed with a low and a high level of interactivity. Big Five personality traits, prior experience with AR or Virtual Reality systems, and various usability data (e.g., subjective ratings related to the usability of heuristics and AR application, number of unique problems, and severity ratings) were collected through questionnaires and utilized during statistical analysis. Results show that even though the employment of the set of GH produced higher overall satisfaction of evaluators, the set of SH resulted in a significantly larger number of unique usability problems. Furthermore, the more interactive version of the AR application was perceived by evaluators as the harder version to inspect, regardless of the set of heuristics utilized. Neuroticism and agreeableness traits were found with a significant impact on subjective usability ratings.

PHH: Policy-Based Hyper-Heuristic With Reinforcement Learning

Hyper-heuristics have a high level of generality and adaptability, allowing them to effectively solve a wide range of complex optimization problems. With reinforcement learning, hyper-heuristics can use experience and knowledge gained to tackle unforeseen problems, allowing hyper-heuristics to adapt and improve over time. Our paper proposes a framework for using policy-based reinforcement learning to improve the performance of hyper-heuristics. The framework trains hyper-heuristic agents to select the best generalized constructive low-level heuristics to solve combinatorial optimization problems. The framework evaluation was performed using three benchmarking problems: traveling salesman, capacitated vehicle routing, and bin packing problems. The results showed that the proposed framework can outperform existing meta-heuristic and hyper-heuristic-based algorithms for all large problem instances in all problem domains. The proposed framework was also evaluated by applying it to a cost optimization problem for workflow scheduling on a hybrid cloud with a deadline constraint. Eight agents were trained on medium-sized workflows with two deadlines and tested against traditional meta-heuristic and hyper-heuristic methods to solve smaller and larger workflows with unforeseen deadlines. Four workflow applications, three workflow sizes, and three deadlines were used in the evaluation. The results showed that our proposed framework provided significantly better solutions of up to 98% for benchmarking problems, and up to 22% for cost optimization in workflow scheduling. Moreover, trained with small problem instances, the framework performed well for unforeseen larger problem instances implying its generalization. The proposed framework, thus, has the potential to improve both the generality and performance of solving large combinatorial optimization problems.

An investigation of F-Race training strategies for cross domain optimisation with memetic algorithms

Parameter tuning is a challenging and time-consuming task, crucial to obtaining improved metaheuristic performance. There is growing interest in cross-domain search methods, which consider a range of optimisation problems rather than being specialised for a single domain. Metaheuristics and hyper-heuristics are typically used as high-level cross-domain search methods, utilising problem-specific low-level heuristics for each problem domain to modify a solution. Such methods have a number of parameters to control their behaviour, whose initial settings can influence their search behaviour significantly. Previous methods in the literature either fix these parameters based on previous experience, or set them specifically for particular problem instances. There is a lack of extensive research investigating the tuning of these parameters systematically. In this paper, F-Race is deployed as an automated cross-domain parameter tuning approach. The parameters of a steady-state memetic algorithm and the low-level heuristics used by this algorithm are tuned across nine single-objective problem domains, using different training strategies and budgets to investigate whether F-Race is capable of effectively tuning parameters for cross-domain search. The empirical results show that the proposed methods manage to find good parameter settings, outperforming many methods from the literature, with different configurations identified as the best depending upon the training approach used.