Search Results (11)

Approximate Computing (AxC) has emerged as a promising paradigm to enhance performance and energy efficiency by allowing a controlled trade-off between accuracy and resource consumption. It is extensively adopted across various abstraction levels, from software to architecture and circuit levels, employing diverse methodologies. The primary objective of AxC is to reduce energy consumption for executing error-resilient applications, accepting controlled and inherently acceptable output quality degradation. However, harnessing AxC poses several challenges, including identifying segments within a design amenable to approximation and selecting suitable AxC techniques to fulfill accuracy and performance criteria. This survey provides a comprehensive review of recent methodologies proposed for performing Design Space Exploration (DSE) to find the most suitable AxC techniques, focusing on both hardware and software implementations. DSE is a crucial design process where system designs are modeled, evaluated, and optimized for various extra-functional system behaviors such as performance, power consumption, energy efficiency, and accuracy. A systematic literature review was conducted to identify papers that ascribe their DSE algorithms, excluding those relying on exhaustive search methods. This survey aims to detail the state-of-the-art DSE methodologies that efficiently select AxC techniques, offering insights into their applicability across different hardware platforms and use-case domains. For this purpose, papers were categorized based on the type of search algorithm used, with Machine Learning (ML) and Evolutionary Algorithms (EAs) being the predominant approaches. Further categorization is based on the target hardware, including Field-Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), general-purpose Central Processing Units (CPUs), and Graphics Processing Units (GPUs). A notable observation was that most studies targeted image processing applications due to their tolerance for accuracy loss. By providing an overview of techniques and methods outlined in existing literature pertaining to the DSE of AxC designs, this survey elucidates the current trends and challenges in optimizing approximate designs. Full article

In this research, considering the air methodology, an axisymmetric model was developed, validated, and calibrated for the numerical simulation of an Oscillating Water Column (OWC) converter subjected to a realistic sea state, representative of the Cassino beach, in the south of Brazil. To do so, the Finite Volume Method (FVM) was used, through the Fluent software (Version 18.1), for the airflow inside the hydro-pneumatic chamber and turbine duct of the OWC. Furthermore, the influence of geometric parameters on the available power of the OWC converter was evaluated through Constructal Design combined with Exhaustive Search. For this, a search space with 100 geometric configurations for the hydro-pneumatic chamber was defined by means of the variation in two degrees of freedom: the ratio between the height and diameter of the hydro-pneumatic chamber (H₁/L₁) and the ratio between the height and diameter of the smallest base of the connection, whose surface of revolution has a trapezoidal shape, between the hydro-pneumatic chamber and the turbine duct (H₂/L₂). The ratio between the height and diameter of the turbine duct (H₃/L₃) was kept constant. The results indicated that the highest available power of the converter was achieved by the lowest values of H₁/L₁ and highest values of H₂/L₂, with the optimal case being obtained by H₁/L₁ = 0.1 and H₂/L₂ = 0.81, achieving a power 839 times greater than the worst case. The values found are impractical in real devices, making it necessary to limit the power of the converters to 500 kW to make this assessment closer to reality; thus, the highest power obtained was 15.5 times greater than that found in the worst case, these values being consistent with other studies developed. As a theoretical recommendation for practical purposes, one can infer that the ratio H₁/L₁ has a greater influence over the OWC’s available power than the ratio H₂/L₂. Full article

Prompt optimization is a crucial task for improving the performance of large language models for downstream tasks. In this paper, a prompt is a sequence of n-grams selected from a vocabulary. Consequently, the aim is to select the optimal prompt concerning a certain performance metric. Prompt optimization can be considered as a combinatorial optimization problem, with the number of possible prompts (i.e., the combinatorial search space) given by the size of the vocabulary (i.e., all the possible n-grams) raised to the power of the length of the prompt. Exhaustive search is impractical; thus, an efficient search strategy is needed. We propose a Bayesian Optimization method performed over a continuous relaxation of the combinatorial search space. Bayesian Optimization is the dominant approach in black-box optimization for its sample efficiency, along with its modular structure and versatility. We use BoTorch, a library for Bayesian Optimization research built on top of PyTorch. Specifically, we focus on Hard Prompt Tuning, which directly searches for an optimal prompt to be added to the text input without requiring access to the Large Language Model, using it as a black-box (such as for GPT-4 which is available as a Model as a Service). Albeit preliminary and based on “vanilla” Bayesian Optimization algorithms, our experiments with RoBERTa as a large language model, on six benchmark datasets, show good performances when compared against other state-of-the-art black-box prompt optimization methods and enable an analysis of the trade-off between the size of the search space, accuracy, and wall-clock time. Full article

In the past decade, the extraction of valuable information from online biomedical datasets has exponentially increased due to the evolution of data processing devices and the utilization of machine learning capabilities to find useful information in these datasets. However, these datasets present a variety of features, dimensionalities, shapes, noise, and heterogeneity. As a result, deriving relevant information remains a problem, since multiple features bottleneck the classification process. Despite their adaptability, current state-of-the-art classifiers have failed to address the problem, giving rise to the exploration of binary optimization algorithms. This study proposes a novel approach to binarizing the Ebola optimization search algorithm. The binary Ebola search optimization algorithm (BEOSA) uses two newly formulated S-shape and V-shape transfer functions to investigate mutations of the infected population in the exploitation and exploration phases, respectively. A model is designed to show a representation of the binary search space and the mapping of the algorithm from the continuous space to the discrete space. Mathematical models are formulated to demonstrate the fitness and cost functions used for evaluating the algorithm. Using 22 benchmark datasets consisting of low, medium and high dimensional data, we exhaustively experimented with the proposed BEOSA method and six other recent similar feature selection methods. The experimental results show that the BEOSA and its variant BIEOSA were highly competitive with different state-of-the-art binary optimization algorithms. A comparative analysis of the classification accuracy obtained for eight binary optimizers showed that BEOSA performed competitively compared to other methods on nine datasets. Evaluation reports on all methods revealed that BEOSA was the top performer, obtaining the best values on eight datasets and eight fitness and cost functions. Computation for the average number of features selected showed that BEOSA outperformed other methods on 11 datasets when population sizes of 75 and 100 were used. Findings from the study revealed that BEOSA is effective in handling the challenge of feature selection in high-dimensional datasets. Full article

Hybridized algorithms are commonly employed to improve the performance of any existing method. However, an optimal learning algorithm composed of evolutionary and swarm intelligence can radically improve the quality of the final neuron states and has not received creative attention yet. Considering this issue, this paper presents a novel metaheuristics algorithm combined with several objectives—introduced as the Hybrid Election Algorithm (HEA)—with great results in solving optimization and combinatorial problems over a binary search space. The core and underpinning ideas of this proposed HEA are inspired by socio-political phenomena, consisting of creative and powerful mechanisms to achieve the optimal result. A non-systematic logical structure can find a better phenomenon in the study of logic programming. In this regard, a non-systematic structure known as Random k Satisfiability (RANkSAT) with higher-order is hosted here to overcome the interpretability and dissimilarity compared to a systematic, logical structure in a Discrete Hopfield Neural Network (DHNN). The novelty of this study is to introduce a new multi-objective Hybrid Election Algorithm that achieves the highest fitness value and can boost the storage capacity of DHNN along with a diversified logical structure embedded with RANkSAT representation. To attain such goals, the proposed algorithm tested four different types of algorithms, such as evolutionary types (Genetic Algorithm (GA)), swarm intelligence types (Artificial Bee Colony algorithm), population-based (traditional Election Algorithm (EA)) and the Exhaustive Search (ES) model. To check the performance of the proposed HEA model, several performance metrics, such as training–testing, energy, similarity analysis and statistical analysis, such as the Friedman test with convergence analysis, have been examined and analyzed. Based on the experimental and statistical results, the proposed HEA model outperformed all the mentioned four models in this research. Full article

In a semi-review paper, we show that the hidden Jahn–Teller effect (JTE) and pseudo-JTE (PJTE) in molecular systems and solids, under certain conditions lead to the formation of two coexisting stable space configurations with different magnetic and dielectric properties, switchable by external perturbations. One of the stable configurations has a high space symmetry and a non-zero or higher spin (HS) (non-zero magnetic moment), the other being distorted, but with zero or lower spin (LS). The number of systems with hidden JTE or PJTE is innumerable; we demonstrate this on the (no exhaustible, too) group of systems with half-filed closed-shell degenerate electronic (orbital or band) configurations e² and t³. The spin-crossover-change from the high symmetry HS arrangement to the low-symmetry LS geometry is accompanied by (driven by the PJTE) orbital disproportionation, in which the system prefers spins-paired states with two electrons on the same orbital (and lower symmetry charge distribution) over the Hund’s spin-parallel arrangement involving several orbitals. Ab initio calculations previously carried out on a series of molecular systems and clusters in crystals, including CuF₃, Si₃, Si₄, Ge₄, C₄H₄, Na₄⁻, C₆₀³⁻, CuO₆ (in two crystal environments, LiCuO₂ and NaCuO₂), etc., confirmed the general theory and allowed for estimates of the parameter values including relaxation times. The hidden JTE and PJTE are thus general tools for search and studies of polyatomic systems with bistabilities. Full article

Deployment of unmanned aerial vehicles (UAVs) as aerial base stations (ABSs) has been considered to be a feasible solution to provide network coverage in scenarios where the conventional terrestrial network is overloaded or inaccessible due to an emergency situation. This article studies the problem of optimal placement of the UAVs as ABSs to enable network connectivity for the users in such a scenario. The main contributions of this work include a less complex approach to optimally position the UAVs and to assign user equipment (UE) to each ABS, such that the total spectral efficiency (TSE) of the network is maximized, while maintaining a minimum QoS requirement for the UEs. The main advantage of the proposed approach is that it only requires the knowledge of UE and ABS locations and statistical channel state information. The optimal 2-dimensional (2D) positions of the ABSs and the UE assignments are found using K-means clustering and a stable marriage approach, considering the characteristics of the air-to-ground propagation channels, the impact of co-channel interference from other ABSs, and the energy constraints of the ABSs. Two approaches are proposed to find the optimal altitudes of the ABSs, using search space constrained exhaustive search and particle swarm optimization (PSO). The numerical results show that the PSO-based approach results in higher TSE compared to the exhaustive search-based approach in dense networks, consuming similar amount of energy for ABS movements. Both approaches lead up to approximately 8-fold energy savings compared to ABS placement using naive exhaustive search. Full article

Building accurate and compact classifiers in real-world applications is one of the crucial tasks in data mining nowadays. In this paper, we propose a new method that can reduce the number of class association rules produced by classical class association rule classifiers, while maintaining an accurate classification model that is comparable to the ones generated by state-of-the-art classification algorithms. More precisely, we propose a new associative classifier that selects “strong” class association rules based on overall coverage of the learning set. The advantage of the proposed classifier is that it generates significantly smaller rules on bigger datasets compared to traditional classifiers while maintaining the classification accuracy. We also discuss how the overall coverage of such classifiers affects their classification accuracy. Performed experiments measuring classification accuracy, number of classification rules and other relevance measures such as precision, recall and f-measure on 12 real-life datasets from the UCI ML repository (Dua, D.; Graff, C. UCI Machine Learning Repository. Irvine, CA: University of California, 2019) show that our method was comparable to 8 other well-known rule-based classification algorithms. It achieved the second-highest average accuracy (84.9%) and the best result in terms of average number of rules among all classification methods. Although not achieving the best results in terms of classification accuracy, our method proved to be producing compact and understandable classifiers by exhaustively searching the entire example space. Full article

Dimensionality reduction research in data envelopment analysis (DEA) has focused on subjective approaches to reduce dimensionality. Such approaches are less useful or attractive in practice because a subjective selection of variables introduces bias. A competing unbiased approach would be to use ensemble DEA scores. This paper illustrates that in addition to unbiased evaluations, the ensemble DEA scores result in unique rankings that have high entropy. Under restrictive assumptions, it is also shown that the ensemble DEA scores are normally distributed. Ensemble models do not require any new modifications to existing DEA objective functions or constraints, and when ensemble scores are normally distributed, returns-to-scale hypothesis testing can be carried out using traditional parametric statistical techniques. Full article

The large number of vehicles constantly seeking access to congested areas in cities means that finding a public parking place is often difficult and causes problems for drivers and citizens alike. In this context, strategies that guide vehicles from one point to another, looking for the most optimal path, are needed. Most contributions in the literature are routing strategies that take into account different criteria to select the optimal route required to find a parking space. This paper aims to identify the types of smart parking systems (SPS) that are available today, as well as investigate the kinds of vehicle detection techniques (VDT) they have and the algorithms or other methods they employ, in order to analyze where the development of these systems is at today. To do this, a survey of 274 publications from January 2012 to December 2019 was conducted. The survey considered four principal features: SPS types reported in the literature, the kinds of VDT used in these SPS, the algorithms or methods they implement, and the stage of development at which they are. Based on a search and extraction of results methodology, this work was able to effectively obtain the current state of the research area. In addition, the exhaustive study of the studies analyzed allowed for a discussion to be established concerning the main difficulties, as well as the gaps and open problems detected for the SPS. The results shown in this study may provide a base for future research on the subject. Full article

“Public legal information from all countries and international institutions is part of the common heritage of humanity. Maximizing access to this information promotes justice and the rule of law”. In accordance with the aforementioned declaration on free access to law by legal information institutes of the world, a plethora of legal information is available through the Internet, while the provision of legal information has never before been easier. Given that law is accessed by a much wider group of people, the majority of whom are not legally trained or qualified, diversification techniques should be employed in the context of legal information retrieval, as to increase user satisfaction. We address the diversification of results in legal search by adopting several state of the art methods from the web search, network analysis and text summarization domains. We provide an exhaustive evaluation of the methods, using a standard dataset from the common law domain that we objectively annotated with relevance judgments for this purpose. Our results: (i) reveal that users receive broader insights across the results they get from a legal information retrieval system; (ii) demonstrate that web search diversification techniques outperform other approaches (e.g., summarization-based, graph-based methods) in the context of legal diversification; and (iii) offer balance boundaries between reinforcing relevant documents or sampling the information space around the legal query. Full article