Most runtime analyses of randomised search heuristics focus on the expected number of function evaluations to find a unique global optimum. We ask a fundamental question: if additional search points are declared optim...
详细信息
ISBN:
(纸本)9781450383523
Most runtime analyses of randomised search heuristics focus on the expected number of function evaluations to find a unique global optimum. We ask a fundamental question: if additional search points are declared optimal, or declared as desirable target points, do these additional optima speed up evolutionary algorithms? More formally, we analyse the expected hitting time of a target set OPT u S where S is a set of non-optimal search points and OPT is the set of optima and compare it to the expected hitting time of OPT. We show that the answer to our question depends on the number and placement of search points in S. For all black-box algorithms and all fitness functions we show that, if additional optima are placed randomly, even an exponential number of optima has a negligible effect on the expected optimisation time. Considering Hamming balls around all global optima gives an easier target for some algorithms and functions and can shift the phase transition with respect to offspring population sizes in the (1,lambda) EA on ONE-MAX. Finally, on functions where search trajectories typically join in a single search point, turning one search point into an optimum drastically reduces the expected optimisation time.
In modern circuit design, highly specialized engineers are using computer tools to increase their chance of finding the best configurations, while decreasing the development time. However, certain tasks, like circuit ...
详细信息
ISBN:
(纸本)9781665427869
In modern circuit design, highly specialized engineers are using computer tools to increase their chance of finding the best configurations, while decreasing the development time. However, certain tasks, like circuit sizing, consist of try and error processes that require the designer's attention for a variable amount of time. The task duration is usually directly proportional to the complexity of the circuit. To minimize the R&D costs of the circuit, relieving the designer from the repetitive tasks is essential. Thus, the trend of replacing manual-based circuit sizing by AI solutions is growing. In this context, we are comparing the five most promising evolutionary algorithms for circuit sizing automation. The focus of this paper is to assess the performance of the algorithms in terms of versatility and population diversity.
This paper presents a simulation-based optimization method for automatic sizing in analog and RF IC blocks. It introduces a combination of a state-of-the-art Multi Objective evolutionary algorithm (EA) with a new cons...
详细信息
ISBN:
(纸本)9781728176703
This paper presents a simulation-based optimization method for automatic sizing in analog and RF IC blocks. It introduces a combination of a state-of-the-art Multi Objective evolutionary algorithm (EA) with a new constraint handling approach to effectively explore the high-dimensional constrained design space, typical in every analog and RF IC block design. An additional modification in the core of the EA is also proposed for handling efficiently mixed continuous-integer parameter search spaces. The methodology is illustrated in a Nested-Current-Mirror amplifier and a Wideband Low Noise Amplifier achieving better results than typical constraint handling approaches.
This paper investigates a comprehensive convolutional neural network (CNN) representation that encodes both layer connections, and computational block attributes for neural architecture search (NAS). We formulate NAS ...
详细信息
ISBN:
(数字)9783030794576
ISBN:
(纸本)9783030794569;9783030794576
This paper investigates a comprehensive convolutional neural network (CNN) representation that encodes both layer connections, and computational block attributes for neural architecture search (NAS). We formulate NAS as a bi-objective optimization problem, where two competing objectives, i.e., the validation accuracy and the model complexity, need to be considered simultaneously. We employ the well-known multi-objective evolutionary algorithm (MOEA) nondominated sorting genetic algorithm II (NSGA-II) to perform multi-objective NAS experiments on the CIFAR-10 dataset. Our NAS runs obtain trade-off fronts of architectures of much wider ranges and better quality compared to NAS runs with less comprehensive representations. We also transfer promising architectures to other datasets, i.e., CIFAR-100, Street View House Numbers, and Intel Image Classification, to verify their applicability. Experimental results indicate that the architectures on the trade-off front obtained at the end of our NAS runs can be straightforwardly employed out of the box without any further modification.
Optimization problems with costly function evaluation widely exist in real-world applications. Surrogate models are commonly used in the field of optimization to deal with such expensive optimization problems. In surr...
详细信息
ISBN:
(纸本)9781728183923
Optimization problems with costly function evaluation widely exist in real-world applications. Surrogate models are commonly used in the field of optimization to deal with such expensive optimization problems. In surrogate model-assisted evolutionary algorithms (EAs), surrogate models like regression models, ranking models or classification models are built based on historical data and then used to compare candidate solutions in place of real function evaluations. Researchers have also proposed various methods to make better use of surrogate models in the optimization process of EAs. However, there is no comprehensive study about how much accuracy of the built model is accurate enough to bring benefits to the optimization. Motivated by this, this work proposes a method to study the performance effect of model accuracy on surrogate model-assisted EAs. Specifically, the method does not really build surrogate models but assumes different model accuracies in individual selection. Two classification-assisted EAs, classification-assisted differential evolution (CADE) and relationship classification-based preselection strategy (RCPS) are analyzed in this work. The experimental results on a set of test functions show that a weak learner with classification accuracy larger than 50% is acceptable ignoring the cost of model building. Another observation is that the performances of CADE and RCPS increase monotonically and nonlinearly with the classification accuracy.
Real-valued genotypes together with the variation operators, mutation and crossover, constitute some of the fundamental building blocks of evolutionary algorithms. Real-valued genotypes are utilized in a broad range o...
详细信息
ISBN:
(纸本)9783030726980;9783030726997
Real-valued genotypes together with the variation operators, mutation and crossover, constitute some of the fundamental building blocks of evolutionary algorithms. Real-valued genotypes are utilized in a broad range of contexts, from weights in Artificial Neural Networks to parameters in robot control systems. Shared between most uses of real-valued genomes is the need for limiting the range of individual parameters to allowable bounds. In this paper we will illustrate the challenge of limiting the parameters of real-valued genomes and analyse the most promising method to properly limit these values. We utilize both empirical as well as benchmark examples to demonstrate the utility of the proposed method and through a literature review show how the insight of this paper could impact other research within the field. The proposed method requires minimal intervention from evolutionary Algorithm practitioners and behaves well under repeated application of variation operators, leading to better theoretical properties as well as significant differences in well-known benchmarks.
Mating restrictions are a mechanism adopted by multi-objective evolutionary algorithms to improve the solution of multi-objective optimization problems (MOPs) by establishing a strategy to mate individuals during the ...
详细信息
ISBN:
(纸本)9781728190488
Mating restrictions are a mechanism adopted by multi-objective evolutionary algorithms to improve the solution of multi-objective optimization problems (MOPs) by establishing a strategy to mate individuals during the reproduction step of the algorithm. Several mating restrictions have been proposed for MOEAs to solve MOPs having two and three objective functions. However, in the case of many-objective optimization problems (four or more objectives), only a few mating restriction schemes have been proposed so far. The Riesz S-energy is a performance indicator which can be used to evaluate population's diversity, and it is able to provide useful neighborhood information from individuals in MOPs with any number of objectives. This feature has been used in some of our previous work to propose a few different mating restriction schemes based on the s-energy indicator. In this paper, we propose the use of an ensemble of four of these mating restriction mechanisms, which is implemented within the NSGA-III to assess its performance. The ensemble's behavior is guided by two measurements of each mating restriction performance throughout the algorithm's execution. We performed an experimental validation of this ensemble in MOPs with up to seven objective functions, and compared the results obtained using the hypervolume, the s-energy, and the inverted generational distance performance indicators. The results obtained show that the use of our mating restrictions ensemble outperforms the original NSGA-III in most of the test instances adopted.
In practise, it is often desirable to provide the decision-maker with a rich set of diverse solutions of decent quality instead of just a single solution. In this paper we study evolutionary diversity optimization for...
详细信息
ISBN:
(纸本)9781450383509
In practise, it is often desirable to provide the decision-maker with a rich set of diverse solutions of decent quality instead of just a single solution. In this paper we study evolutionary diversity optimization for the knapsack problem (KP). Our goal is to evolve a population of solutions that all have a profit of at least (1- epsilon).OPT, where OPT is the value of an optimal solution. Furthermore, they should differ in structure with respect to an entropy-based diversity measure. To this end we propose a simple (mu + 1)-EA with initial approximate solutions calculated by a well-known FPTAS for the KP. We investigate the effect of different standard mutation operators and introduce biased mutation and crossover which puts strong probability on flipping bits of low and/or high frequency within the population. An experimental study on different instances and settings shows that the proposed mutation operators in most cases perform slightly inferior in the long term, but show strong benefits if the number of function evaluations is severely limited.
State-of-the-art Convolutional Neural Networks (CNNs) have become increasingly accurate. However, hundreds or thousands of megabytes data are involved to store them, making these networks also computationally expensiv...
详细信息
ISBN:
(纸本)9781728190488
State-of-the-art Convolutional Neural Networks (CNNs) have become increasingly accurate. However, hundreds or thousands of megabytes data are involved to store them, making these networks also computationally expensive. For certain applications, such as Internet-of-Things (IoT), where such CNNs are to be implemented on resource-constrained and memory-constrained platforms, including Field-Programmable Gate Arrays (FPGAs) and embedded devices, CNN architectures and parameters have to be small and efficient. In this paper, an evolutionary algorithm (EA) based adaptive integer quantisation method is proposed to reduce network size. The proposed method uses single objective rank-based evolutionary strategy to find the best quantisation bin boundary for fixed quantised bit width. The performance of the proposed method is evaluated on a small CNN, the LeNet-5 architecture, using the CIFAR-10 dataset. The aim is to devise a methodology that allows adaptive quantisation of both weights and bias from 32-bit floating point to 8-bit integer representation for LeNet-5, while retaining accuracy. The experiments compare straight-forward (linear) quantisation from 32-bits to 8-bits with the proposed adaptive quantisation method. The results show that the proposed method is capable of quantising CNNs to lower bit width representation with only a slight loss in classification accuracy.
In this work we present a pattern classification approach coupling the Neighbourhood Component Analysis (NCA) classifier with the Canonical Differential evolutionary Particle Swarm Optimization (C-DEEPSO). The standar...
详细信息
ISBN:
(纸本)9781728183923
In this work we present a pattern classification approach coupling the Neighbourhood Component Analysis (NCA) classifier with the Canonical Differential evolutionary Particle Swarm Optimization (C-DEEPSO). The standard NCA uses the conjugate gradient method to minimize the classification error. Here we propose an approach using the C-DEEPSO instead. In the experimental design, the coupled approach is applied to 20 benchmark data sets, and its performance is compared with the standard NCA using the conjugate gradient. The experimental analysis shows the usage of an evolutionary approach to enhance the performance of a machine learning algorithm can be competitive when compared to well-known iterative optimization techniques, and even outperform them in some problems. A real-world problem classifying cyber-attacks to an industrial control system of gas pipelines is also solved by the proposed approach. The results obtained indicate the proposed approach can successfully identify possible cyber-attacks to the control system. In this way, the NCA coupled to C-DEEPSO can work as an Intrusion Detection Systems (IDS), being able to guarantee an acceptable security level.
暂无评论