Complex Search Research Articles

Dual Elite Groups-Guided Differential Evolution for Global Numerical Optimization

Differential evolution (DE) has shown remarkable performance in solving continuous optimization problems. However, its optimization performance still encounters limitations when confronted with complex optimization problems with lots of local regions. To address this issue, this paper proposes a dual elite groups-guided mutation strategy called “DE/current-to-duelite/1” for DE. As a result, a novel DE variant called DEGGDE is developed. Instead of only using the elites in the current population to direct the evolution of all individuals, DEGGDE additionally maintains an archive to store the obsolete parent individuals and then assembles the elites in both the current population and the archive to guide the mutation of all individuals. In this way, the diversity of the guiding exemplars in the mutation is expectedly promoted. With the guidance of these diverse elites, a good balance between exploration of the complex search space and exploitation of the found promising regions is hopefully maintained in DEGGDE. As a result, DEGGDE expectedly achieves good optimization performance in solving complex optimization problems. A large number of experiments are conducted on the CEC’2017 benchmark set with three different dimension sizes to demonstrate the effectiveness of DEGGDE. Experimental results have confirmed that DEGGDE performs competitively with or even significantly better than eleven state-of-the-art and representative DE variants.

Where to meet on heat? A conceptual framework for optimising demand reduction and decarbonised heat supply

The question of how to optimally decarbonise the buildings sector – especially heating – remains a source of fierce debate. A large amount of uncertainty remains around how to strike the optimal balance between heat demand reduction and heat supply decarbonisation. This paper sheds light on the complex search for the optimal balance between reducing heat demand and decarbonising heat supply. Drawing on economic theory and cost-benefit analysis, the paper proposes an analytical framework that structures and organises the many complex and intertwining parameters to consider when determining this optimal balance. We conclude that increased clarity about the goal, scope of, and method for optimisation efforts are the foundations of a more informed and considered approach when it comes to decarbonising heating. The paper also provides insights for policy makers faced with the challenge of designing policies and regulations in this area.

Don't hide the instruction manual: A dynamic trade-off between using internal and external templates during visual search.

Visual search is typically studied by requiring participants to memorize a template initially, for which they subsequently search in a crowded display. Search in daily life, however, often involves templates that remain accessible externally, and may therefore be (re)attended for just-in-time encoding or to refresh internal template representations. Here, we show that participants indeed use external templates during search when given the chance. This behavior was observed during both simple and complex search, scaled with task difficulty, and was associated with improved performance. Furthermore, we show that participants used external sampling not only to offload memory, but also as a means of verifying whether the template was remembered correctly at the end of trials. We conclude that the external world may not only provide the challenge (e.g., distractors), but may dynamically ease search. These results argue for extensions of state-of-the-art models of search, because external sampling seems to be used frequently, in at least two ways and is actually beneficial for task performance. Our findings support a model of visual working memory that emphasizes a resource-efficient trade-off between storing and (re)attending external information.

Сравнение и отбор ситуаций в системах вывода решений на прецедентах для «умной» фермы

The trend of development of smart farms is aimed at their becoming fully autonomous, robotic enterprises. The prospects for the intellectualization of agricultural production and smart farms, in particular, today are associated with the development of technology systems used to detect, recognize complex production situations and search for effective solutions in these situations. The article presents the concept of such a decision support system on smart farms using the method of decision support based on case-based reasoning - CBR system. Its implementation requires a number of non-trivial tasks, which include, first of all, the tasks of formalizing the presentation of situations and creating methods for comparing and retrieving situations from the KB on this basis. In this study, a smart farm is presented as a complex technological object consisting of interrelated components, which are the technological subsystems of a smart farm, the products produced, the objects of the operational environment, as well as the relationships between them. To implement algorithms for situational decision-making based on precedents, a formalized representation of the situation in the form of a multivector is proposed. This allowed us to develop a number of models of the trained similarity function between situations. The conducted experiments have shown the operability of the proposed models, on the basis of which ensemble architecture of a neural network has been developed for comparing situations and selecting them from the knowledge base in decision-making processes. Of practical interest is monitoring the condition of plants by their video and photo images, which allows detecting undesirable plant conditions (diseases), which can serve as a signal to activate the process of searching for solutions in the knowledge base.

GtAI: an improved species-specific tRNA adaptation index using the genetic algorithm.

The tRNA adaptation index (tAI) is a translation efficiency metric that considers weighted values (S ij values) for codon-tRNA wobble interaction efficiencies. The initial implementation of the tAI had significant flaws. For instance, generated S ij weights were optimized based on gene expression in Saccharomyces cerevisiae, which is expected to vary among different species. Consequently, a species-specific approach (stAI) was developed to overcome those limitations. However, the stAI method employed a hill climbing algorithm to optimize the S ij weights, which is not ideal for obtaining the best set of S ij weights because it could struggle to find the global maximum given a complex search space, even after using different starting positions. In addition, it did not perform well in computing the tAI of fungal genomes in comparison with the original implementation. We developed a novel approach named genetic tAI (gtAI) implemented as a Python package (https://github.com/AliYoussef96/gtAI), which employs a genetic algorithm to obtain the best set of S ij weights and follows a new codon usage-based workflow that better computes the tAI of genomes from the three domains of life. The gtAI has significantly improved the correlation with the codon adaptation index (CAI) and the prediction of protein abundance (empirical data) compared to the stAI.

Sine cosine algorithm with communication and quality enhancement: Performance design for engineering problems

Abstract In recent years, the sine cosine algorithm (SCA) has become one of the popular swarm intelligence algorithms due to its simple and convenient structure. However, the standard SCA tends to fall into the local optimum when solving complex multimodal tasks, leading to unsatisfactory results. Therefore, this study presents the SCA with communication and quality enhancement, called CCEQSCA. The proposed algorithm includes two enhancement strategies: the communication and collaboration strategy (CC) and the quality enhancement strategy (EQ). In the proposed algorithm, CC strengthens the connection of SCA populations by guiding the search agents closer to the range of optimal solutions. EQ improves the quality of candidate solutions to enhance the exploitation of the algorithm. Furthermore, EQ can explore potential candidate solutions in other scopes, thus strengthening the ability of the algorithm to prevent trapping in the local optimum. To verify the capability of CCEQSCA, 30 functions from the IEEE CEC2017 are analyzed. The proposed algorithm is compared with 5 advanced original algorithms and 10 advanced variants. The outcomes indicate that it is dominant over other comparison algorithms in global optimization tasks. The work in this paper is also utilized to tackle three typical engineering design problems with excellent optimization capabilities. It has been experimentally demonstrated that CCEQSCA works as an effective tool to tackle real issues with constraints and complex search space.

EPPFM: Efficient and Privacy-Preserving Querying of Electronic Medical Records With Forward Privacy in Multiuser Setting

With the application of the Internet of Things (IoT) and cloud computing, the eHealthcare industry has developed markedly, attracting many patients to seek medical treatment in an eHealthcare system. However, for patients who first register in the system, due to lack of experience, an important aspect is to choose appropriate medical services. Considering the sensitivity of health care data and the semi-honest nature of the cloud server, it is a good solution to use searchable encryption (SE) to obtain some historical electronic medical records (EMRs) that are consistent with the patient's symptom keyword combination and have high service scores for reference. However, existing SE schemes still have issues meeting the requirements of the eHealthcare system for flexible authorization and revocation, efficiency, and forward privacy. To resolve these issues, we propose two efficient and privacy-preserving electronic medical records query schemes with forward privacy in a multiuser setting (EPPFM). First, we present the basic scheme EPPFM-I to achieve a multiuser multikeyword exact match query under linear search complexity. In EPPFM-I, we also use the pseudorandom function (PRF) to perform the function of forward privacy. Then, we use a bucket structure to construct the improved scheme EPPFM-II, which has a faster-than-linear search complexity. Finally, we use detailed security analysis and extensive simulations to show the security and efficiency of the proposed schemes, respectively.

Heuristic Search for Physics-Based Problems: Angry Birds in PDDL+

This paper studies how a domain-independent planner and combinatorial search can be employed to play AngryBirds, a well established AI challenge problem. To model the game, we use PDDL+, a planning language for mixed discrete/continuous domains that supports durative processes and exogenous events. The paper describes the PDDL+ model and identifies key design decisions that reduce the problem complexity. In addition, we propose several domain-specific enhancements including heuristics and a search technique similar to preferred operators. Together, they alleviate the complexity of combinatorial search. We evaluate our approach by comparing its performance with dedicated domain-specific solvers on a range of Angry Birds levels. The results show that our performance is on par with these domain-specific approaches in most levels, even without using our domain-specific search enhancements.

Graphs, Constraints, and Search for the Abstraction and Reasoning Corpus

The Abstraction and Reasoning Corpus (ARC) aims at benchmarking the performance of general artificial intelligence algorithms. The ARC's focus on broad generalization and few-shot learning has made it difficult to solve using pure machine learning. A more promising approach has been to perform program synthesis within an appropriately designed Domain Specific Language (DSL). However, these too have seen limited success. We propose Abstract Reasoning with Graph Abstractions (ARGA), a new object-centric framework that first represents images using graphs and then performs a search for a correct program in a DSL that is based on the abstracted graph space. The complexity of this combinatorial search is tamed through the use of constraint acquisition, state hashing, and Tabu search. An extensive set of experiments demonstrates the promise of ARGA in tackling some of the complicated object-centric tasks of the ARC rather efficiently, producing programs that are correct and easy to understand.

Dynamic Heterogeneous Graph Attention Neural Architecture Search

Dynamic heterogeneous graph neural networks (DHGNNs) have been shown to be effective in handling the ubiquitous dynamic heterogeneous graphs. However, the existing DHGNNs are hand-designed, requiring extensive human efforts and failing to adapt to diverse dynamic heterogeneous graph scenarios. In this paper, we propose to automate the design of DHGNN, which faces two major challenges: 1) how to design the search space to jointly consider the spatial-temporal dependencies and heterogeneous interactions in graphs; 2) how to design an efficient search algorithm in the potentially large and complex search space. To tackle these challenges, we propose a novel Dynamic Heterogeneous Graph Attention Search (DHGAS) method. Our proposed method can automatically discover the optimal DHGNN architecture and adapt to various dynamic heterogeneous graph scenarios without human guidance. In particular, we first propose a unified dynamic heterogeneous graph attention (DHGA) framework, which enables each node to jointly attend its heterogeneous and dynamic neighbors. Based on the framework, we design a localization space to determine where the attention should be applied and a parameterization space to determine how the attention should be parameterized. Lastly, we design a multi-stage differentiable search algorithm to efficiently explore the search space. Extensive experiments on real-world dynamic heterogeneous graph datasets demonstrate that our proposed method significantly outperforms state-of-the-art baselines for tasks including link prediction, node classification and node regression. To the best of our knowledge, DHGAS is the first dynamic heterogeneous graph neural architecture search method.

A universal large-scale many-objective optimization framework based on cultural learning

When solving large-scale many-objective optimization problems (LMaOPs), due to the large number of variables and objectives involved, the algorithm is faced with a very high-dimensional and complex search space, which is difficult to be explored with limited resources. To address these issues, this paper proposes a universal large-scale many-objective optimization framework based on cultural learning (UCLMO). First, a universal framework is proposed, and multi-objective optimizers can be embedded into the framework to accelerate the convergence. Moreover, inspired by cultural learning, an individual selection strategy based on historical knowledge is proposed to promote the diversity of the population, and an assisted evolution strategy based on normative knowledge is presented to accelerate the convergence of the algorithm. Experiments have been conducted on multi-objective knapsack problems and LMaOPs with decision variables ranging from 500 to 1500, and the number of objectives ranging from 5 to 15. The experimental results verify the superiority and competitiveness of the proposed UCLMO framework in solving LMaOPs compared with state-of-the-art algorithms.

Online tourist information search strategies

Technological advances continue to significantly change tourist information search behaviour. This paper revisits online tourist information search strategies by reconceptualising them as a combination of goals, approaches and online sources. Based on pre-trip and post-trip interviews and trip planning diaries, the paper investigates how tourists search for information online. An online tourist information search strategy framework is developed by considering how tourists integrate approaches and sources to meet their information search goals. The key contributions of the study are the redefinition of tourist information search strategy in an online context, a classification of sources that more suitably reflects online information-scapes, and the identification of interesting and often complex search patterns that suggest changes in information search strategies over the course of trip planning.

An Attribute-Based Searchable Encryption Scheme for Cloud-Assisted IIoT

The searchable encryption is a particular case of structured encryption, which has been intensively researched in the secure cloud storage system. By constructing structured secure index such as encrypted multi-maps (EMM), encrypted inverted index (EII), etc., searchable encryption can achieve efficient keyword search over encrypted data set. However, existing searchable encryption constructions do not take search permissions into consideration, resulting in the lack of a mechanism of the data access control, which may not be suitable for IIoT applications, since an integrated industrial system contains all kinds of data with rigorous access permissions. In this paper, we construct an attribute-based searchable encryption construction for cloud-assisted IIoT application scenario. By designing the novel access policy based structured secure index and the attribute-based search token, our construction achieves fine-grained keyword search privilege control over encrypted IIoT data as well as the same search complexity as the traditional searchable encryption. To the best of our knowledge, this is the first Attribute-Based Searchable Encryption (ABSE) construction. We provide the correctness and security proofs for our construction. Experimental evaluation results in a real-world data set show the correctness and the practical search efficiency of the proposed ABSE.

The supply of information and price formation: Evidence from Google's search engine

AbstractThis study develops several Google search‐based measures to test the relation between earnings‐week online search results and the speed of price discovery. These measures are based on searches using only a firm's ticker symbol in the search string. I collect the total number of search results (across all search result pages) as well as the type and content of search results on the first three pages of search results. I find that the quantity, quality, and content of search results have varying effects on the speed at which earnings news is impounded into stock price. I also find that effects are only observed for search results presented on the first page of a Google search. Overall, my results suggest that (1) increases in online information are associated with slower price discovery, and (2) the likely mechanism by which this association operates is through useful search results being crowded off the first page of results by more complex or irrelevant search results.

Joint-optimized feature selection and classifier hyperparameters by salp swarm algorithm in piano score difficulty measurement problem

This study proposes the semantic-explicit features that characterize difficulty, and jointly optimizes feature selection and classifier hyperparameters by the salp swarm algorithm (SSA) to deal with the corresponding mixed-integer programming problem with constructing large-scale piano score difficulty level datasets. The difficulty level of piano scores is essential for piano learners to choose the appropriate piece, especially for beginners and amateurs. However, the previous studies lack an open-access baseline dataset and sufficient difficulty-related features, as well as the separate optimization of and feature and model hyperparameter. To address such problems, this study constructs large-scale difficulty-level datasets, proposes novel difficulty-related features, and jointly optimizes feature selection and classifier hyperparameters due to the coupled effect of feature selection and model optimization. The search space of the joint optimization is complex due to the strong mutual constraint relationship of difficulty levels in the piano-score difficulty measurement (PSDM) problem. SSA is adapted to the joint optimization scheme of the PSDM problem with the advantages of only one main controlling parameter and less computation complexity involving the gradual SSA movement approach to balance global exploration and local exploitation in an unknown and complex search space. The joint-optimization mechanism by SSA achieves an overall accuracy of 78.80% and 60.68% on two datasets of 677 and 2040 piano pieces with difficulty levels of four and nine, respectively. The results of recognition accuracy obviously validate the distinguished performance of our joint-optimization scheme compared to the successive optimization and joint optimization by other seven optimization algorithms in terms of the PSDM problem.

Intelligently Wireless Batteryless RF-Powered Reconfigurable Surface: Theory, Implementation & Limitations

This work exploits ultra-low cost, commodity, radio frequency identification (RFID) tags as elements of a reconfigurable intelligent surface (RIS). Such batteryless tags are powered and controlled by a software-defined radio reader, with properly modified software, so that a source-destination link is assisted, operating at a different band. Signal model includes small-scale and large-scale fading, direct link, as well as specific parameters relevant to reflection (i.e., backscatter) radio, such as antenna structural mode and reflection efficiency, typically overlooked in the literature. An algorithm is offered that computes the optimal RIS configuration with complexity of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">O</i> ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">M</i> log <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">M</i> ) in number of elements <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">M</i> , instead of intractable exponential complexity of exhaustive search, while accommodating any number <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> ≥ 2 of loads. With the proposed algorithm, it is shown that performance gains reach a plateau for constant element spacing and increasing number of elements, suggesting that the weak, passive nature of backscattered links limits the performance gains, even with perfect channel estimation. Channel estimation with linear minimum mean squared error (LMMSE) estimator is shown to be effective, provided that there are sufficient number of pilot symbols. A concrete way is offered to design and prototype a wireless, batteryless, RF-powered, reconfigurable surface and a proof-of-concept is experimentally demonstrated.

A bibliometric analysis of immuno-oncology research: Is a new global leader emerging?

e13593 Background: The last decade has witnessed an increase in cancer research globally, and a transformation of the anti-cancer armamentarium. The biggest innovation has been development of immuno-oncology (IO) therapies, and specifically immune checkpoint inhibitors (ICIs), targeting CTLA-4 and PD-(L)1, reshaping treatment paradigms for many cancers (e.g., melanoma). Since the first ICI approval in 2011, research from countries in the European Union (EU) and North America has dominated this field. However, more recently research authored in China has emerged as a potential leader. In this study we analysed cancer and IO research outputs since 2011, exploring emerging national trends, and focusing on ICI development. Methods: IO research articles were identified using a high-resolution bibliometric method, previously validated. Relevant articles were identified from the Web of Science, utilising a complex search strategy of pre-defined keywords, including the 9 FDA-approved ICI and 8 ICI approved in China-only. Additional search filters were used to determine secondary characteristics including country (assigned a fractional count based on proportional authorship), collaborations, and tumour site. Results: Since 2011, over 175,000 cancer research articles have been published globally. The number of articles published annually has increased steadily, largely driven by increasing outputs from Chinese authors. China is now the leading global publisher of cancer research accounting for 18% of all outputs in 2021, and has consistently surpassed US and EU total outputs since 2019. Over the past decade there has been a steady increase in total IO research outputs, rising by 378% (n = 2926) compared to 2011. China began publishing significant IO research in 2014 and has risen rapidly to dominate this field. In 2021, China accounted for 37% of total IO outputs. The US was previously dominant, however there has been a steady decline in proportional outputs from 63% in 2012-12 to 29% in 2019-21. Only 17% of published Chinese IO research is internationally collaborative. Further, Chinese research focuses on tumour sites with a high national prevalence, such as oesophagus, stomach and liver, dominating research outputs in these areas. Research on the 9 FDA-approved ICI is dominated by the US, with only 7% having Chinese author contribution. Conversely, the 8 ICIs approved in China-only have negligible (&lt; 5%) author contributions from the US and EU. Conclusions: The past decade has witnessed a substantial rise in both cancer and IO research. IO articles published by Chinese authors, frequently exclusively, now account for over a third of annual outputs. Those authored in China focus on cancers with a high national prevalence, and link to ICIs approved exclusively in China. With increasing global appetite for ICIs, especially from middle-income countries, ICIs developed in China may offer a viable alternative to current options.

Neural Image Parts Group Search for Person Re-Identification

Employing partition strategy to explore fine-grained features has been verified to be beneficial for person re-identification in recent literature. However, existing methods primarily rely on expert experience to manually design various partition strategies, which may lead to a sub-optimal solution for fine-grained features exploration. In this paper, we propose a Neural Parts Group Search (NPGS) strategy that auto-searches the optimal parts group via evolutionary algorithm (EA) to facilitate the network to exploit the local details. And during search process, designing a high-quality search space is especially crucial for an efficient optimization. Considering the human top-down structure and the semantic coherence of parts, we design a coarse-to-fine parts search space (C2F-PSP) in NPGS, which effectively reduce the search complexity without the loss of parts expressivity. Additionally, since only employing the high-level semantic features is insufficient for the NPGS to search effective parts, we further develop an efficient feature aggregation strategy named hierarchical low-rank bilinear pooling that progressively integrates the high-level semantic property and the low-level fine-grained details to facilitate the NPGS to explore the fine-grained features. Furthermore, to relieve the interference of background during parts search process, we propose a novel Relational Attention Module (RAM) by exploiting the channel and spatial structural interdependence of pixels to strengthen the discriminative regions. Extensive experiments on the mainstream evaluation datasets demonstrate that our method outperforms the recent state-of-the-art Re-ID models.

Efficient Strong Privacy-Preserving Conjunctive Keyword Search Over Encrypted Cloud Data

Searchable symmetric encryption (SSE) supports keyword search over outsourced symmetrically encrypted data. Dynamic searchable symmetric encryption (DSSE), a variant of SSE, further enables data updating. Most DSSE works with conjunctive keyword search primarily consider forward and backward privacy. Ideally, the server should only learn the result sets involving all keywords in the conjunction. However, existing schemes suffer from keyword pair result pattern (KPRP) leakage, revealing the partial result sets containing two of query keywords. We propose the first DSSE scheme to address aforementioned concerns that achieves strong privacy-preserving conjunctive keyword search. Specifically, our scheme can maintain forward and backward privacy and eliminate KPRP leakage, offering a higher level of security. The search complexity scales with the number of documents stored in the database in several existing schemes. However, the complexity of our scheme scales with the update frequency of the least frequent keyword in the conjunction, which is much smaller than the size of the entire database. Besides, we devise a least frequent keyword acquisition protocol to reduce frequent interactions between clients. Finally, we analyze the security of our scheme and evaluate its performance theoretically and experimentally. The results show that our scheme has strong privacy preservation and efficiency.

Sleep-wakeup scheduling algorithm for lifespan maximization of directional sensor networks: a discrete cuckoo search optimization algorithm

Directional sensor networks (DSNs) are ad-hoc networks which are utilized in different industrial applications. Their usual engagements are to monitor and to perform the coverage of all specific targets in the observing fields permanently. These kinds of networks include numerous configurable directional sensors in which they can be utilized in one of the possible directions along with the one of their adjustable ranges. Although the energy harvesting methodology is being applied for these battery-hungry applications, the battery management and network lifetime maximization are still prominent challenges. In this paper, the network lifetime extension is formulated to a discrete optimization problem which is a famous non-deterministic polynomial time hardness (NP-Hard) problem. To solve this combinatorial problem, a discrete cuckoo search algorithm (D-CSA) is designed and is called in several rounds. A cover is a sub set of configured sensors capable of monitoring all targets in the observing field. In each round, the most efficient cover is constituted along with its activation time. In the determined activation time, the sensors in the cover are scheduled in wakeup mode whereas others are set in sleep mode to save energy. Despite other meta-heuristic algorithms, this proposed algorithm utilizes the novel defined discrete walking around procedures that makes to reach a good balance between exploration and exploitation in this complex search space. The proposed algorithm has been tested in different scenarios to be evaluated. The simulation results in the variety circumstances prove the superiority of the proposed algorithm is about 20.29%, 19.55%, 14.40%, 14.51%, 7.70% and 8.03% in term of average lifespan improvement against H-MNLAR, Hm-LifMax-BC, GA, ACOSC, H-GATS, and HDPSO algorithms, respectively. The results also show the high potential scalability of the proposed algorithm.