Accurate blood glucose prediction is vital for effective diabetes management, yet the complex, non-linear nature of glycemic dynamics poses a significant challenge for evolutionary regression models. Standard selection mechanisms, which typically rely on aggregated fitness metrics (e.g., global RMSE), often suffer from premature convergence, leading to the stagnation of the search process in local optima. This paper investigates the application of Desire Driven Selection (DDS) within Structured Grammatical Evolution (SGE) for glucose forecasting. Inspired by biological sexual selection, DDS decouples reproduction from strict survival pressure. It allows individuals to co-evolve mating preferences, selecting partners based on "ornaments" that represent performance during specific time-of-day windows rather than a single global score. We evaluated this approach using real-world clinical data from 10 patients with diabetes. Experimental results indicate that SGE+DDS significantly enhances the search capability of the algorithm, outperforming standard Tournament Selection in generalization ability for 9 out of 10 patients. Notably, the proposed method achieved reductions in prediction error of up to 19.5% compared to the baseline on the test data. Analysis reveals that while DDS introduces higher variance between runs compared to the Tournament selection, it increased exploratory capacity, discovering high-quality models that capture difficult glycemic fluctuations.

Diabetes clinicians want glucose forecasting models that are, at the same time, precise and easy to understand. Creating these models is very difficult due to the complex dynamics of glucose in people with diabetes. Grammatical Evolution (GE) is a promising method for forecasting glucose levels, as it creates clear mathematical models that are both accurate and easy to interpret. However, as with other evolutionary algorithms, the performance of GE is linked to the fitness landscape, which has not been sufficiently studied.

The formation of nitrogen oxides (NOx) in combustion systems is notable for its harmful impact on public health and the environment. Therefore, it is imperative to develop models to predict NOx formation in different situations. These models are designed to capture the characteristics of three distinct engine operating states: nominal, startup, and saturation. The nominal state represents the typical operating conditions, the startup state refers to the initial phase of the operation of the engine, and the saturation state corresponds to the operation of the engine at its maximum capacity. We applied dynamic structured grammatical evolution to obtain a set of interpretable expressions, which are mathematical representations capable of capturing the dynamics of NOx formation in combustion systems and that can be easily interpreted. These models were compared with traditional differential equation-based models to assess their interpretability and predictive accuracy for the three scenarios. Through our approach, we obtained a set of interpretable expressions that improved those obtained by a differential equation-based mathematical model, providing a more transparent and intuitive understanding of the system's behavior. Our technique seeks to unveil the dynamics of NOx formation processes that could significantly reduce NOx emissions and mitigate their impact on global environmental pollution.

People with Type 1 diabetes need to predict their blood glucose levels regularly to keep them within a safe range. Accurate predictions help prevent short-term issues like hypoglycemia and reduce the risk of long-term complications. Evolutionary algorithms have shown potential for this task by generating reliable models for glucose prediction.

ABSTRACT Receiver operating characteristic curves are commonly used to evaluate the performance of machine learning ensemble classification models that combine multiple classifiers through a voting procedure. Although these models have many parameters, standard ROC analyses typically vary only the voting threshold, limiting their potential for improvement. In this paper, we propose Performance Curve Mapping, a new method that redefines the ROC curve as the Pareto front of a multi‐objective optimisation problem. The method maps the multidimensional space of all ensemble parameters (Decision space) into a two‐dimensional Objective space defined by classification performance metrics. We employ an algorithm based on NSGA‐II to explore the Decision space and validate the proposal on two different classification problems: (1) predicting car insurance claims in a highly imbalanced dataset (Insurance dataset), and (2) predicting obesity risk in a balanced clinical dataset (GenObIA dataset). We compare our method with alternative ensemble optimisation approaches, using visual assessment, the area under the curve and the Youden index as performance measures. In the Insurance dataset, Performance Curve Mapping achieves an average improvement of 46.4% in AUC‐ROC and 26.1% in the Youden index. In the GenObIA dataset, it achieves an average improvement of 29.7% in AUC‐ROC and 11.9% in the Youden index. All improvements are calculated relative to the maximum achievable improvement.

Modern AI applications contain computationally expensive sections. Accelerator cards and tools like AMD Vitis HLS leverage high-level synthesis and hardware (HW) optimizations to create custom HW designs to accelerate them.Nevertheless, the learning curve is steep, even for those with previous knowledge of HW design, due to the complexity of the optimization techniques and limited information on their interactions and HW effects. This paper quantitatively analyzes the interactions of optimization techniques after reverse engineering Vitis’ optimization directives, both in isolation and in pairs.Over 150 experiments were conducted to investigate three distinct goals: assessing pragma behavior and the rules governing pragma application and optimizations, modeling Vitis HLS latency estimates, and evaluating the impact of optimizations on design space exploration, specifically area and latency. These experiments involve different combinations and placements of optimizations in the loop and function hierarchy of the test bench. Our findings offer guidance on using Vitis pragmas and identify promising configurations for optimizing latency and area.

This article proposes the first hardware implementation of a low-power LSTM neural network targeting a wearable medical device designed to predict blood glucose at a 30-minute horizon. This work aims to reduce energy consumption by proposing new activation functions that target hardware implementation. On top of this proposal, we also prove there is room for improvement in energy consumption by applying neural network optimizations at the algorithmic, such as quantization, and architecture level, LSTM hyperparameters, that consider the target hardware. To validate our proposal, we devise an optimized version of the neural network aimed to be wearable and, therefore, to reduce its energy consumption while preserving its accuracy as much as possible. The hardware is implemented on a Xilinx Virtex-7 FPGA VC707 Evaluation Kit. It is compared with (i) a faithful design of the original neural network implemented on the same evaluation kit, (ii) three state-of-the-art LSTM-based FPGA implementations, and (iii) software implementations running in cutting-edge smartphones: OnePlus Nord and an Apple iPhone 13 Pro with artificial intelligence hardware accelerators. Our proposal consumes between ×1020 and ×7 less energy than the software implementations, being the most efficient system compared to the smartphones. On the other hand, its energy efficiency, measured in GFLOP/J, is between ×$2.84$ and ×$7.82$ greater than other state-of-the-art LSTM implementations, proving to be the most suitable implementation for a wearable system for blood glucose prediction.

People with diabetes must carefully monitor their blood glucose levels, especially after eating. Blood glucose management requires a proper combination of food intake and insulin boluses. Glucose prediction is vital to avoid dangerous post-meal complications in treating individuals with diabetes. Although traditional methods, and also artificial neural networks, have shown high accuracy rates, sometimes they are not suitable for developing personalised treatments by physicians due to their lack of interpretability. This study proposes a novel glucose prediction method emphasising interpretability: Interpretable Sparse Identification by Grammatical Evolution. Combined with a previous clustering stage, our approach provides finite difference equations to predict postprandial glucose levels up to two hours after meals. We divide the dataset into four-hour segments and perform clustering based on blood glucose values for the two-hour window before the meal. Prediction models are trained for each cluster for the two-hour windows after meals, allowing predictions in 15-minute steps, yielding up to eight predictions at different time horizons. Prediction safety was evaluated based on Parkes Error Grid regions. Our technique produces safe predictions through explainable expressions, avoiding zones D (0.2% average) and E (0%) and reducing predictions on zone C (6.2%). In addition, our proposal has slightly better accuracy than other techniques, including sparse identification of non-linear dynamics and artificial neural networks. The results demonstrate that our proposal provides interpretable solutions without sacrificing prediction accuracy, offering a promising approach to glucose prediction in diabetes management that balances accuracy, interpretability, and computational efficiency.

People with diabetes have to properly manage their blood glucose levels in order to avoid acute complications. This is a difficult task and an accurate and timely prediction may be of vital importance, specially of extreme values. Perhaps one of the main concerns of people with diabetes is to suffer an hypoglycemia (low value) event and moreover, that the event will be prolonged in time. It is crucial to predict events of hyperglycemia (high value) and hypoglycemia that may cause health damages in the short term and potential permanent damages in the long term. The aim of this paper is to describe our research on predicting hypoglycemia events using Dynamic structured Grammatical Evolution. Our proposal gives white box models induced by a grammar based on if-then-else conditions. We trained and tested our system with real data collected from 5 different diabetic patients, producing 30 minutes predictions with encouraging results.

Blood glucose management is a difficult task that people with diabetes usually have to perform by themselves. An accurate and timely prediction is vital in order to take decisions and recommend corrective actions to the patient when the future blood glucose value lies outside of a target range. It is crucial to predict events like hyperglycemia and hypoglycemia that may cause health damages in the short term and potential permanent damages in the long term. The aim of this paper is to describe our research on predicting blood glucose values using a two phase Takagi-Sugeno-Kang Fuzzy Rule Based System. The first phase is a learning process where membership functions and rules are optimised by a genetic algorithm. In the second phase of tuning, we used a genetic algorithm to perform the selection and optimisation of the rules. To train our model we used two different scenarios, What-if and Agnostic; in both of them the inputs are values measured by a continuous monitoring glucose system as well as previous carbohydrate intake and insulin injections. In the What-if scenario, assumed future values of meals and insulin injections are permitted. On the other hand, in the Agnostic scenario, only information of the past and present events are available for the prediction. We trained and tested our system with real data collected from 10 different diabetic patients, producing 30, 60, 90 and 120 minutes predictions with encouraging accuracy results.

Type 1 Diabetes patients have to control their blood glucose levels using insulin therapy. Numerous factors (such as carbohydrate intake, physical activity, time of day, etc.) greatly complicate this task. In this article we propose a modeling method that will allow us to make predictions of blood glucose level evolution with a time horizon of 24 hours. This may allow the adjustment of insulin doses in advance and could help to improve the living conditions of diabetes patients. Our approach starts from a system of finite difference equations that characterizes the interaction between insulin and glucose (in the field, this is known as a minimal model). This model has several parameters whose values vary widely depending on patient characteristics and time. Thus, in the first phase of our strategy, We will enrich the patient's historical data by adding white Gaussian noise, which will allow us to perform a probabilistic fitting with a 95% confidence interval. Then, the model's parameters are adjusted based on the history of each patient using a genetic algorithm and dividing the day into 12 time intervals. In the final stage, we will perform a whole-day forecast from an ensemble of the models fitted in the previous phase. The validity of our strategy will be tested using the Parkers' error grid analysis. Our experimental results based on data from real diabetic patients show that this technique is capable of robust predictions that take into account all the uncertainty associated with the interaction between insulin and glucose.

Energy is essential for all countries, since it is in the core of social and economic development. Since the industrial revolution, the demand for energy has increased exponentially. It is expected that the energy consumption in the world increases by 50% by 2030 [17]. As such, managing the demand of energy is of the uttermost importance. The development of tools to model and accurately predict the demand of energy is very important to policy makers. In this paper we propose the use of the Structured Grammatical Evolution (SGE) algorithm to evolve models of energy demand, over macro-economic indicators. The proposed SGE is hybridised with a Differential Evolution approach in order to obtain the parameters of the models evolved which better fit the real energy demand. We have tested the performance of the proposed approach in a problem of total energy demand estimation in Spain, where we show that the SGE is able to generate extremely accurate and robust models for the energy prediction within one year time-horizon.

Abstract In this paper, we analyze the population diversity of grammatical evolution (GE) on multiple levels of genetic information: chromosome diversity, expression diversity, and output diversity. Thereby, we use a tree-similarity metric from tree-based GP literature to determine similarity of expression trees generated in GE. The similarity of outputs is determined via their correlation. We track the pairwise similarities for all individuals within a generation on all three levels and track the distribution of similarity values over generations. We demonstrate the analysis method using four symbolic regression problem instances and find that the visualization highlights some issues that can occur when using GE such as: large groups of individuals with highly similar outputs, a high fraction of trees with constant outputs, or short and highly similar trees in the early stages of the GE run. Especially in the early phases of GE, we see that a large subset of the population represents equivalent expressions. In early stages, rather short expressions are produced leaving large parts of the chromosome unexpressed. More complex expressions can be derived only after GE has successfully evolved well-working beginnings of chromosomes.

This article presents an approach to providing fault tolerance to permanent effects in the substrate of dynamic, partially reconfigurable field-programmable gate arrays (FPGAs). Our proposal consists of modifying FPGA configuration at runtime to avoid permanently damaged regions of the FPGA. It demands that the circuit design fulfills several requirements regarding the functionality and interfaces of its reconfigurable modules, the structure of the communications, the inclusion of specialized modules to handle and optimize circuit reconfiguration at runtime, and the organization of the reconfigurable partitions. We evaluate two methods to select the new target partition for modules in faulty partitions and design their hardware as intellectual property modules. We evaluate the performance and scalability of this approach using a trapezoidal shaper, a filter used to detect high-energy particles in radiation experiments, and we carry out the experiments with a variable number of modules and reconfigurable partitions using the two selection algorithms. The proposal is compared with nonfault-tolerant and triple modular redundancy approaches, and it remains functional with up to 12× more injected faults than those.

Structured grammatical evolution is a recent grammar-based genetic programming variant that tackles the main drawbacks of Grammatical Evolution, by relying on a one-to-one mapping between each gene and a non-terminal symbol of the grammar. It was applied, with success, in previous works with a set of classical benchmarks problems. However, assessing performance on hard real-world problems is still missing. In this paper, we fill in this gap, by analyzing the performance of SGE when generating predictive models for the glucose levels of diabetic patients. Our algorithm uses features that take into account the past glucose values, insulin injections, and the amount of carbohydrate ingested by a patient. The results show that SGE can evolve models that can predict the glucose more accurately when compared with previous grammar-based approaches used for the same problem. Additionally, we also show that the models tend to be more robust, since the behavior in the training and test data is very similar, with a small variance.

Diabetes Mellitus Type 1 patients are waiting for the arrival of the Artificial Pancreas. Artificial Pancreas systems will control the blood glucose of patients, improving their quality of life and reducing the risks they face daily. At the core of the Artificial Pancreas, an algorithm will forecast future glucose levels and estimate insulin bolus sizes. Grammatical Evolution has been proved as a suitable algorithm for predicting glucose levels. Nevertheless, one of the main obstacles that researches have found for training the Grammatical Evolution models is the lack of significant amounts of data. As in many other fields in medicine, the collection of data from real patients is very complex along with the fact that the patient's response can vary in a high degree due to a lot of personal factors which can be seen as different scenarios. In this paper, we propose both a classification system for scenario selection and a data augmentation algorithm that generates synthetic glucose time series from real data. Our experimental results show that, in a scarce data context, Grammatical Evolution models can get more accurate and robust predictions using scenario selection and data augmentation.

Different techniques have been proposed to tackle GA-Hard problems. Some techniques work with different encodings and representations, other use reordering operators and several, such as the Evolutionary Mapping Method (EMM), apply genotype-phenotype mappings. EMM uses multiple chromosomes in a single cell for mating with another cell within a single population. Although EMM gave good results, it fails on solving some deceptive problems. In this line, EMMRS (EMM with Replacement and Shift) adds a new operator, consisting on doing a replacement and a shift of some of the bits within the chromosome. Results showed the efficiency of the proposal on deceptive problems. However, EMMRS was not tested with other kind of hard problems. In this paper we have adapted EMMRS for solving the Traveling Salesman Problem (TSP). The encodings and genetic operators for solving the TSP are quite different to those applied on deceptive problems. In addition, execution times recommended the parallelization of the GA. We implemented a GPU parallel version. We present here some preliminary results proving that Evolutionary Mapping Method with Replacement and Shift gives good results not only in terms of quality but also in terms of speedup on its GPU parallel version for some instances of the TSP problem.

This special issue includes the extended version of selected papers presented at the fourth Parallel Architectures and Bioinspired Algorithms Workshop held in Galveston Island (TX, USA) on October 14, 2011 in conjunction withParallel Architectures and Compilation Techniques (PACT). This workshop follows the success of the three previous workshops held in conjunction with PACT 2008 in Toronto, Canada 1, PACT 2009 in Raleigh, USA 2, and PACT 2010 in Vienna 3, Austria and the two previous Workshops on Parallel Bioinspired Algorithms 4 held in Oslo, 2005, (together with IEEE International Conference on Parallel Processing (ICPP)) 5 and London, 2007 (together with Association for Computing Machinery (ACM) Genetic and Evolutionary Computation Conference (Gecco) 2007) 6. These series of workshops has shown that knowledge fields such as parallel computer architectures and Parallel and Distributed Computing and Bioinspired Algorithms, which could seem quite different in a first approach, are able to find transversal elements that enrich them. Bioinspired Algorithms comprise a set of heuristics that can help to optimize a wide range of problems, including many tasks faced by parallel architectures designers, such as balancing computer load, fault-tolerance and dependability, thermal-aware design, and NoC design. In addition, Bioinspired Algorithms may help in finding solutions related to compilation and resource sharing issues, which are interesting problems for parallel architectures. Parallel architecture designers may also propose infrastructures that allow the improvement of computing performance of Bioinspired Algorithms, which usually face real-world problems that manage huge amounts of data. Alternatives to the classical sequential solution are needed by this community. Therefore, topics such as P2P, cluster and grid computing, cloud computing, and Graphics Processing Unit (GPUs) implementation of Bioinspired Algorithms are very interesting to this field. We have therefore considered this opportunity to give a broader view on the application of nature inspired computing techniques to hardware design and parallel architectures problem solving. We have also considered the interest of including an overview of the available bioinspired techniques based tools that have been used so far to solve problems related to automatic hardware design. We thus invited Professors Oscar Garnica and Juan Lanchares to work with us with this aim. Therefore, this special issue includes the paper entitled A review of bioinspired CAD tools for parallel architectures and hardware design 7 together with four papers carefully selected for publications, extended versions of Parallel Architectures and Bioinspired Algorithms 2011 workshop's best papers. The first paper selected, GPU-based acceleration of bio-inspired motion estimation model 8, describes the specific and efficient implementation of a gradient-based optical flow model. The proposed model enhances the GPU computing capability when compared with other optical flow gradient family algorithms and has been particularized using a validated neuromorphic motion estimation system for the robust extraction of image velocity. The second paper, entitled Evaluation of asynchronous multi-swarm particle optimization on several topologies 9, evaluates the impact of the topology on multi-swarm systems, considering that swarms are independently interacting only when particle migration occurs. Several topologies and communication strategies have been evaluated within the paper, including broadcast and gossip on fully connected networks, unidirectional and bidirectional rings, hypercubes, and a dynamic topology. The work KLONOS: Similarity-based planning tool support for porting scientific applications 10 proposes a methodology to address planning support, an important aspect of software porting that usually receives little attention. When porting a scientific application, the selection of key subroutines greatly impacts the productivity . The authors propose ad methodology on the basis of the idea that a set of similar subroutines can be ported with similar strategies and result in a similar-quality porting. They apply bioinformatics techniques to conduct the similarity analysis of subroutines, by viewing subroutines as data and operator sequences, analogous to DNA sequences. Finally, the work entitled Boosting the 3D thermal-aware floorplanning problem through a master-worker parallel MOEA 11 deals with the problem of placing the hardware components on a 3D chip to reduce overheating by dissipation. The main contribution of this paper is to present a parallelization of the evaluation phase in a master-worker model to achieve a dramatic speed-up of the thermal-aware floorplanning process. Exhaustive experimentation was carried out over 3D integrated circuits, with 48 and 128 cores, outperforming previous published works. This work has been partially supported by Spanish Government grants Avanza Competitividad I+D+i TSI-020100-2010-962, TIN 2008-00508, MEC Consolider Ingenio CSD00C-07-20811, TIN2011-28627-C04-03, and GRU10029 Gobierno de Extremadura and EDRF, and Municipality of Almendralejo.

SUMMARY During the tools have been also influenced by this evolutionary fashion. In this paper, we give a broader view of the application of techniques inspired by nature to hardware design and parallel architectures problem solving. Our aim is to furnish an overview of the various bioinspired techniques based tools that have been used so far to solve the problems of automatic hardware design. We can claim that a lot of the approaches found in the literature suffer from a lack of interdisciplinary interaction among researchers of both evolutionary computation and hardware design fields. In addition, we have also detected that some multi‐objective problems do not use the appropriate algorithms. Copyright © 2012 John Wiley & Sons, Ltd.

In the context of Parallel Evolutionary Algorithms, it has been shown that different population structures induce different search performances. Nevertheless, no work has shown a clear cut evidence that there is a correlation between the solver’s population structure and the problem’s network structure. In this work, we verify this correlation performing a clear and systematic analysis of a large set of population structures (based on the well known β -graphs and NK -landscape problems. Furthermore, we go beyond our findings in these idealised experiments by analysing the performance of variable-topology EAs on a dynamic real-world problem, the Multi-Skills Call Centre.

Technical indicators (TIs) are used to interpret stock market and to predict market trends. The main difficulty in the use of TIs lies in deciding which their optimal parameter values are in each moment, since constant optimal values do not seem to exist. In this work, the use of Multi-Objective Evolutionary Algorithms (MOEAs) is proposed to obtain the best values of the parameters in order to help to buy and sell shares. Those parameters are applied in real time and belong to a collection of indicators. Unlike other previous approaches, the necessity of repeating the parameter optimization process each time a new data enters the system is justified, searching for the best adjustment of the parameters (and hence the TIs) in every moment. The Moving Averages Convergence-Divergence (MACD) indicator and the Relative Strength Index (RSI) oscillator have been chosen as TIs, so the MOEAs will provide the best parameters to use them on investment decisions. Experiments compare up to nine different configurations with the Buy & Hold strategy (B & H). The obtained results show that the Multi-Objective technique proposed here can greatly improve the results of the B & H strategy even operating daily. This statement is also demonstrated by comparing the results to those previously presented in the literature.

The dynamic memory manager (DMM) is a key element whose customization for a target application reports great benefits in terms of execution time, memory usage and energy consumption. Previous works presented algorithms to automatically obtain custom DMMs for a given application. Nevertheless, those approaches are based on grammatical evolution where the fitness is built as an aggregate objective function, which does not completely exploit the search space, returning the designer the DMM solution with best fitness. However, this approach may not find solutions that could fit in a concrete hardware platform due to a very low value of one of the objectives while the others remain high, which may represent a high fitness. In this work we present the first multi-objective optimization methodology applied to DMM optimization where the Pareto dominance is considered, thus providing the designer with a set of non-dominated DMM implementations on each optimization run. Our results show that the multi-objective optimization provides Pareto-optimal alternatives due to a better exploitation of the search space obtaining better hypervolume values than the aggregate objective function approach.

Heat removal and power density distribution delivery have become two major reliability concerns in 3D stacked technology. Additionally, the placement of Through-Silicon-Vias (TSVs) for connecting different layers is one of the key issues in 3D technology. Although a few recent works have considered thermal-aware placement of cores in chip multi-processor architectures, the concepts of 3D and TSVs have not been conveniently incorporated. Therefore, new suitable exploration methods for the 3D thermal-aware floorplaning problem need to be developed. In this paper we analyze the benefits of two different exploration techniques for the floorplanning problem: Multi-Objective Genetic Algorithm (MOGA) and a Mixed Integer Linear Program (MILP). We present a novel algorithm that uses MILP to minimize average temperature in the 3D chip, whereas uses MOGA to insert TSVs, connecting the layers while the total wire length is minimized. Our experiments with two different 3D chips show that our algorithm achieves 10% reduction in the maximum temperature and thermal gradient.

This paper presents a Named Entity Recognition (NER) system based on Hidden Markov Models. The system design is language independent, and the target language and scope of the NER is determined by the training corpus. The NER is formed by two subsystems that detect and label the entities independently. Each subsystem implements a different approach of that statistical theory, showing that each component may complement the results of the other one. Unlike most of the previous works, two labels are returned when the components provide different results. This redundancy is an advantage when human supervision is mandatory at the end of the process such as in intelligence environments.

This paper investigates the use of empirical and Archimedean copulas as probabilistic models of continuous estimation of distribution algorithms (EDAs). A method for learning and sampling empirical bivariate copulas to be used in the context of n-dimensional EDAs is first introduced. Then, by using Archimedean copulas instead of empirical makes possible to construct n-dimensional copulas with the same purpose. Both copula-based EDAs are compared to other known continuous EDAs on a set of 24 functions and different number of variables. Experimental results show that the proposed copula-based EDAs achieve a better behaviour than previous approaches in a 20% of the benchmark functions.

Resizable caches can trade-off capacity for access speed to dynamically match the needs of the workload. In Simultaneous Multi-Threaded (SMT) cores, the caching needs can vary greatly across the number of threads and their characteristics, offering opportunities to dynamically adjust cache resources to the workload. In this paper we propose the use of resizable caches in order to improve the performance of SMT cores, and introduce a new control algorithm that provides good results independent of the number of running threads. In workloads with a single thread, the resizable cache control algorithm should optimize for cache miss behavior because misses typically form the critical path. In contrast, with several independent threads running, we show that optimizing for cache hit behavior has more impact, since large SMT workloads have other threads to run during a cache miss. Moreover, we demonstrate that these seemingly diametrically opposed policies can be simultaneously satisfied by using the harmonic mean of the per-thread speedups as the metric to evaluate the system performance, and to smoothly and naturally adjust to the degree of multithreading.

Current general-purpose memory allocators do not provide sufficient speed or flexibility for modern high-performance applications. To optimize metrics like performance, memory usage and energy consumption, software engineers often write custom allocators from scratch, which is a difficult and error-prone process. In this paper, we present a flexible and efficient simulator to study Dynamic Memory Managers (DMMs), a composition of one or more memory allocators. This novel approach allows programmers to simulate custom and general DMMs, which can be composed without incurring any additional runtime overhead or additional programming cost. We show that this infrastructure simplifies DMM construction, mainly because the target application does not need to be compiled every time a new DMM must be evaluated. Within a search procedure, the system designer can choose the "best" allocator by simulation for a particular target application. In our evaluation, we show that our scheme will deliver better performance, less memory usage and less energy consumption than single memory allocators.

Technology scaling has offered advantages to embedded systems, such as increased performance, more available memory and reduced energy consumption. However, scaling also brings a number of problems like reliability degradation mechanisms. The intensive activity of devices and high operating temperatures are key factors for reliability degradation in latest technology nodes. Focusing on embedded systems, the memory is prone to suffer reliability problems due to the intensive use of dynamic memory on wireless and multimedia applications. In this work we present a new approach to automatically design dynamic memory managers considering reliability, and improving performance, memory footprint and energy consumption. Our approach, based on Grammatical Evolution, obtains a maximum improvement of 39% in execution time, 38% in memory usage and 50% in energy consumption over state-of-the-art dynamic memory managers for several real-life applications. In addition, the resulting distributions of memory accesses improve reliability. To the best of our knowledge, this is the first proposal for automatic dynamic memory manager design that considers reliability.

Thermal effects are becoming increasingly important in today's sub-micron technologies. Thermal issues affect the performance, the reliability and the cooling costs of integrated systems. High peak temperatures are of major concern in modern 3D designs, where the stacking of multiple layers leads to higher power densities. Therefore, the integration of the thermal-aware design during the initial phases of the design can reduce the cost and the time-to-market of the resulting product. An efficient floorplanning in terms of thermal effects will reduce the appearance of critical hotspots and will spread heat across the chip area.

New portable consumer embedded devices must execute multimedia applications (e.g., 3D games, video players and signal processing software, etc.) that demand extensive memory accesses and memory usage at a low energy consumption. Moreover, they must heavily rely on Dynamic Memory (DM) due to the unpredictability of the input data and system behavior. Within this context, consistent design methodologies that can tackle efficiently the complex DM behavior of these multimedia applications are in great need. In this article, we present a novel design framework, based on genetic programming, which allows us to design custom DM management mechanisms, optimizing memory accesses, memory use and energy consumption for the target embedded system. First, we describe the large design space of DM management decisions for multimedia embedded applications. Then, we propose a suitable way to traverse this design space using grammatical evolution and construct custom DM managers that minimize the DM used by these highly dynamic applications. As a result, our methodology achieves significant improvements in memory accesses (23% less on average), memory usage (38% less on average) and energy consumption (reductions of 21% on average) in real case studies over the current state-of-the-art DM managers used for these types of dynamic applications. To the best of our knowledge, this is the first approach to efficiently design DM managers for embedded systems using evolutionary computation and grammar evolution.

New multimedia embedded applications are becoming increasingly dynamic. Thus, they cannot only rely on static data allocation, and must employ Dynamically-allocated Data Types (DDTs) to store their data and efficiently use the limited physical resources of embedded devices. However, the optimization of the DDTs for each target embedded system is a very time-consuming process due to the large design space of possible DDTs implementations and selection for the memory hierarchy of each specific embedded device. Thus, new suitable exploration methods for embedded design metrics (memory accesses, usage and power consumption) need to be developed. In this paper we analyze the benefits of two different exploration techniques for DDTs optimization: Multi-Objective Particle Swarm Optimization (MOPSO) and a Mixed Integer Linear Program (MILP). Furthermore, we propose a novel MOPSO exploration method, OMOPSO*, which uses MILP solutions, as reference points, to guide a MOPSO exploration and reach solutions closer to the real Pareto front of solutions. Our experiments with two real-life embedded applications show that our algorithm achieves 40% better coverage and set of solutions than state-of-the-art optimization methods for DDTs (MOGAs and other MOPSOs).

ESL

New multimedia embedded applications are increasingly dynamic, and rely on Dynamically-allocated Data Types (DDTs) to store their data. The optimization of DDTs for each target embedded system is a time-consuming process due to the large design space of possible DDTs implementations. Thus, suitable exploration methods for embedded design metrics (memory accesses, memory usage and power consumption) need to be developed. In this work we present a detailed analysis of the characteristics of different types of Multi-Objective Evolutionary Algorithms (MOEAs) to tackle the optimization of DDTs in multimedia applications and compare them with other state-of-the-art heuristics. Our results with state-of-the-art MOEAs in two object-oriented multimedia embedded applications show that more sophisticated MOEAs (SPEA2 and NSGA-II) offer better solutions than simple schemes (VEGA). Moreover, the suitable sophisticated scheme varies according to the available exploration time, namely, NSGA-II outperforms SPEA2 in the first set of solutions (300-500 generations), while SPEA2 offers better solutions afterwards.

This paper presents a new method for solving discrete deceptive problems using a genotype to phenotype mapping where a new replacement and shift operator is applied. The method is evaluated using different deceptive problems. Experimental results show how our method obtains a speed-up of 94% with respect to other approaches.

Asynchronous systems are attracting the interest of a growing number of designers. However, the lack of simulation tools devoted to asynchronous microarchitectures is a gap that is not narrowed today. One of the main obstacles on the simulation of asynchronous systems is the variable computation delays of their modules, which compute as fast as possible under the actual conditions of the system because there is no clock signal. In this paper we present a modelling method that describes the variable computation delay of an asynchronous circuit by using probability distribution functions that return the probability of a given delay to be spent on the computation of a data. This method was integrated in an architectural simulator of a 64-bit superscalar asynchronous microarchitecture where the computation delay of each one of the modules of the microarchitecture was characterized through a probability distribution function. The experimental results showed that the asynchronous behavior was successfully modeled, and the architectural simulations of standard benchmarks were affordable in terms of wall-clock simulation time.

This paper presents a research about the Fault Tolerancenature of the Island Model when applied to Distributed-Parallel Genetic Algorithms (GAs). Parallel and distributed models have been extensively applied to GAs when researchers tackle hard problems. Nevertheless, there are few works dealing with the problem of failures that are usually present when a distributed infrastructure is employed. The main results from this research suggest that the GAs Island Modelsare fault tolerant by nature.

Embedded consumer devices are increasing their capabilities and can now implement new multimedia applications reserved only for powerful desktops a few years ago. These applications share complex and intensive dynamic memory use. Thus, dynamic memory optimizations are a requirement when porting these applications. Within these optimizations, the refinement of the Dynamically (de)allocated Data Type (or DDT) implementations is one of the most important and difficult parts for an efficient mapping onto low-power embedded devices. In this paper, we describe a new automatic optimization approach for the DDTs of object-oriented multimedia applications. It is based on an analytical pre-characterization of the possible elementary DDT blocks, and a multi-objective genetic algorithm to explore the design space and to select the best implementation according to different optimization criteria (i.e., memory accesses, memory footprint and energy consumption). Our results in real-life multimedia applications show that the best implementations of DDTs can be obtained in an automated way in few hours, while typically designers would require days to find a suitable implementation, achieving important savings in exploration time with respect to other state-of-the-art heuristics-based optimization methods for this task.

Nowadays, synchronous processor designers have to deal with severe problems related to the distribution of a complex clock network like skew reduction, high power-consumption, synchronization of clocks, etc. Asynchronous or self-timed architectures are becoming an interesting design alternative because they usually avoid these drawbacks, and they are able to achieve high performance at a low power consumption cost. However, on the first steps of the design process, the evaluation of the performance of such architectures through simulations is much more complicated due to the requirement of modeling the data-dependant timing of each system module. The aim of this paper is to evaluate the performance of a 64-bit fully-asynchronous superscalar processor microarchitecture with dynamically scheduled instruction flow, out-of-order speculative execution of instructions and advanced branch prediction. To tackle this goal we have described the asynchronous microarchitecture solving the synchronization between structures through a four-phase handshake protocol. Then, we have used a modification of the SimpleScalar suite to model the asynchronous microarchitecture in order to run Alpha programs on it. Finally, we have compared the performance of this fully-asynchronous processor with the performance obtained from its synchronous counterpart by running architectural simulations of the SPEC2000 benchmarks on both models

This paper presents a hardware technique to reduce the static and dynamic power consumption in functional units of a 64-bit superscalar processor. Our approach is based on substituting some of the 64-bit power-hungry adders by others with 32-bit lower power-consumption adders, and modifying the protocol in order to issue as much instructions as possible to those low power-consumption units incurring a negligible performance penalty. Our technique saves between 14.7% and a 50% of the power-consumption in the adders which is between 6.1% and a 20% of power-consumption in the execution units. This reduction is important because it can avoid the creation of a hot spot on the functional units

It is usually difficult to find a balance among some of the important parameters when using an evolutionary algorithm (EA) (number of runs, population size and generations) and at the same time saving computing time. Recently, some papers have dealt with population size and optimal numbers of populations, while others have instead focused on a different couple of parameters, and scarcely the three parameters have been considered simultaneously. In this paper we consider simultaneously all of them. Computing effort is used through experimental results section to evaluate the proposed alternatives. Experimental results confirm some conclusions obtained on previous works with only two parameters and also give some guidelines on the way of distributing efficiently resources when designing parallel implementations of EAs.

Genetic algorithms (GAs) are stochastic optimization heuristics in which searches in solution space are carried out by imitating the population genetics stated in Darwin's theory of evolution. The compact genetic algorithm (cGA) does not manage a population of solutions but only mimics its existence. The combination of genetic and local search heuristic has been shown to be an effective approach to solve some optimization problems more efficiently than with a single GA or a cGA. multi-FPGA systems design flow has three major tasks: partitioning, placement and routing. In this paper we present a new hybrid algorithm that exploits a cGA in order to generate high quality partitioning and placement solutions and, by means of a local search heuristic, improves the solutions obtained using a cGA or a GA.

One of the goals of a high level synthesis process is to minimize the circuit implementation cost. Since the minimization problem associated with those transformations is NP complete, in this work we present an evolutionary algorithm that optimize circuit specifications by means of a special type of genetic operator. We have named this operator algebraic mutation, carried out with the help of algebraic equations. This work can be classified within the algebraic optimization of equational specifications of circuits by using genetic techniques. We have applied this technique to a simple circuit equational specification and to a much more complex algebraic equation. In the first case our algorithm simplifies the equation until the optimum specification is found and in the second a solution improving the former is always obtained, and when we increase the population size, the optimum solution is also found.

One of the main bottlenecks in Genetics Algorithms is the fitness evaluation for each individual. In this work, we propose a new fitness evaluation method, which will solve the bottleneck calculating a fitness on the fly.

The combination of genetic and local search heuristics has been shown to be an effective approach to solving the traveling salesman problem (TSP). This paper describes a new hybrid algorithm that exploits a compact genetic algorithm in order to generate high-quality tours, which are then refined by means of the Lin-Kernighan (LK) local search. The local optima found by the LK local search are in turn exploited by the evolutionary part of the algorithm in order to improve the quality of its simulated population. The results of several experiments conducted on different TSP instances with up to 13,509 cities show the efficacy of the symbiosis between the two heuristics.