检索结果-内蒙古大学图书馆

An evolutionary algorithm with crossover and mutation for model-based clustering

学校读者我要写书评

暂无评论

arXiv 2018年

作者： McNicholas, Sharon M. McNicholas, Paul D. Ashlock, Daniel A. Department of Mathematics and Statistics McMaster University ON Canada Department of Mathematics and Statistics University of Guelph ON Canada

The expectation-maximization (EM) algorithm is almost ubiquitous for parameter estimation in model-based clustering problems;however, it can become stuck at local maxima, due to its single path, monotonic nature. Rather than using an EM algorithm, an evolutionary algorithm (EA) is developed. This EA facilitates a different search of the fitness landscape, i.e., the likelihood surface, utilizing both crossover and mutation. Furthermore, this EA represents an efficient approach to "hard" model-based clustering and so it can be viewed as a sort of generalization of the k-means algorithm, which is itself equivalent to a classification EM algorithm for a Gaussian mixture model with spherical component covariances. The EA is illustrated on several data sets, and its performance is compared to k-means clustering as well as model-based clustering with an EM algorithm. Copyright © 2018, The Authors. All rights reserved.

关键词： evolutionary algorithms

Supervising Unsupervised Learning with evolutionary Algorithm in Deep Neural Network

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Inagaki, Takeshi IBM Tokyo Japan

A method to control results of gradient descent unsupervised learning in a deep neural network by using evolutionary algorithm is proposed. To process crossover of unsupervisedly trained models, the algorithm evaluates pointwise fitness of individual nodes in neural network. Labeled training data is randomly sampled and breeding process selects nodes by calculating degree of their consistency on different sets of sampled data. This method supervises unsupervised training by evolutionary process. We also introduce modified Restricted Boltzmann Machine which contains repulsive force among nodes in a neural network and it contributes to isolate network nodes each other to avoid accidental degeneration of nodes by evolutionary process. These new methods are applied to document classification problem and it results better accuracy than a traditional fully supervised classifier implemented with linear regression algorithm. Copyright © 2018, The Authors. All rights reserved.

关键词： evolutionary algorithms

Preselection via one-class classification for evolutionary optimization 18

学校读者我要写书评

暂无评论

Preselection via one-class classification for evolutionary o...

Proceedings of the Genetic and evolutionary Computation Conference Companion

作者： Jinyuan Zhang Aimin Zhou Guixu Zhang East China Normal University Shanghai China

ISBN: (纸本)9781450357647

In evolutionary algorithms, a preselection operator aims to choose some promising offspring solutions for a further environmental selection. Most existing preselection operators are based on fitness values, surrogate models, or classification models. Since a preselection operation can be regarded as a classification procedure, the classification based preselection is a natural choice for evolutionary algorithms. However, it is not trivial to prepare 'positive' and 'negative' training samples by using binary and/or multi-class classification models in preselection. To deal with this problem, this paper proposes a one-class classification based preselection (OCPS) scheme, which only needs one class of 'positive' training samples. The proposed OCPS scheme is applied to two state-of-the-art evolutionary algorithms on a test suite. The experimental results show the potential of OCPS on improving the performance of some existing evolutionary algorithms.

关键词： one-class classification evolutionary algorithms preselection

A Comparison of the taguchi method and evolutionary optimization in multivariate testing

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Jiang, Jingbo Legrand, Diego Severn, Robert Miikkulainen, Risto Sentient Technologies Inc. One California California St Suite 2300 San FranciscoCA94111 United States University of Texas at Austin 2317 Speedway Stop D9500 AustinTX78712 United States

Multivariate testing has recently emerged as a promising technique in web interface design. In contrast to the standard A/B testing, multivariate approach aims at evaluating a large number of values in a few key variables systematically. The Taguchi method is a practical implementation of this idea, focusing on orthogonal combinations of values. This paper evaluates an alternative method: population-based search, i.e. evolutionary optimization. Its performance is compared to that of the Taguchi method in several simulated conditions, including an orthogonal one designed to favor the Taguchi method, and two realistic conditions with dependences between variables. evolutionary optimization is found to perform significantly better especially in the realistic conditions, suggesting that it forms a good approach for web interface design in the future. Copyright © 2018, The Authors. All rights reserved.

关键词： evolutionary algorithms

A Simple Decomposition-Based Many-Objective evolutionary Algorithm with Local Iterative Update

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Zhang, Yingyu Zeng, Bing School of Computer Science Liaocheng University Liaocheng252059 China School of Software Engineering South China University of Technology Guangzhou510006 China

Existing studies have shown that the conventional multi-objective evolutionary algorithms (MOEAs) based on decomposition may lose the population diversity when solving some many-objective optimization problems. In this paper, a simple decompositionbased MOEA with local iterative update (LIU) is proposed. The LIU strategy has two features that are expected to drive the population to approximate the Pareto Front with good distribution. One is that only the worst solution in the current neighborhood is swapped out by the newly generated offspring, preventing the population from being occupied by copies of a few individuals. The other is that its iterative process helps to assign better solutions to subproblems, which is beneficial to make full use of the similarity of solutions to neighboring subproblems and explore local areas in the search space. In addition, the time complexity of the proposed algorithm is the same as that of MOEA/D, and lower than that of other known MOEAs, since it considers only individualswithin the current neighborhood at each update. The algorithmis comparedwith several of the bestMOEAs on problems chosen from two famous test suites DTLZ and WFG. Experimental results demonstrate that only a handful of running instances of the algorithm on DTLZ4 lose their population diversity. What's more, the algorithm wins in most of the test instances in terms of both running time and solution quality, indicating that it is very effective in solving MaOPs. Copyright © 2018, The Authors. All rights reserved.

关键词： evolutionary algorithms

Discrepancy-based evolutionary Diversity Optimization

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Neumann, Aneta Gao, Wanru Doerr, Carola Neumann, Frank Wagner, Markus Optimisation and Logistics University of Adelaide Adelaide Australia CNRS Sorbonne Universités UPMC Univ Paris 06 LIP6 Paris France

Diversity plays a crucial role in evolutionary computation. While diversity has been mainly used to prevent the population of an evolutionary algorithm from premature convergence, the use of evolutionary algorithms to obtain a diverse set of solutions has gained increasing attention in recent years. Diversity optimization in terms of features on the underlying problem allows to obtain a better understanding of possible solutions to the problem at hand and can be used for algorithm selection when dealing with combinatorial optimization problems such as the Traveling Salesperson Problem. We explore the use of the star-discrepancy measure to guide the diversity optimization process of an evolutionary algorithm. In our experimental investigations, we consider our discrepancy-based diversity optimization approaches for evolving diverse sets of images as well as instances of the Traveling Salesperson problem where a local search is not able to find near optimal solutions. Our experimental investigations comparing three diversity optimization approaches show that a discrepancy-based diversity optimization approach using a tie-breaking rule based on weighted differences to surrounding feature points provides the best results in terms of the star discrepancy measure. Copyright © 2018, The Authors. All rights reserved.

关键词： evolutionary algorithms

How do simple evolutionary strategies and investment optimizations a ect ecological patterns? the case of generalized Taylor's Law

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Garlaschi, Stefano Stivanello, Samuele Department of Physics and Astronomy Galileo Galilei University of Padova Department of Mathematics Tullio Levi-Civita University of Padova

Taylor's Law (TL) relates the variance to the mean of a random variable via power law. In ecology it applies to populations and it is a common empirical pattern shared among di erent ecosystems. Measurements give power law exponent to be between 1 and 2, and more often to cluster around 2, whereas theoretical models predict TL exponent can assume any real value. In this paper, adopting the framework of multiplicative growth models in a Markovian environment, we investigate the possibility of evolutionary strategies to be responsible for TL exponent to be in a finite range. We implement three di erent strategies the individuals can follow and for each strategy set two di erent optimization investment objectives. In all the studied cases we find TL exponent can assume any real value due to the existence of regions of the model parameters in which the exponent can diverge. Furthermore, under natural hypothesis on the dynamics of the environment, the shapes of these regions do not depend on di erent strategies adopted and nor on the optimization objective. Thus the introduction of strategies dose not a ect the range of TL exponent in the model. In our theoretical framework rare events are shaping the value of the TL exponent, suggesting, as hinted by previous works, that empirical values may be a statistical artifact following from under sampling. Copyright © 2018, The Authors. All rights reserved.

关键词： evolutionary algorithms

Guided evolutionary Strategies: Escaping the curse of dimensionality in random search

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Maheswaranathan, Niru Metz, Luke Tucker, George Choi, Dami Sohl-Dickstein, Jascha Google Brain

Many applications in machine learning require optimizing a function whose true gradient is unknown, but where surrogate gradient information (directions that may be correlated with, but not necessarily identical to, the true gradient) is available instead. This arises when an approximate gradient is easier to compute than the full gradient (e.g. in meta-learning or unrolled optimization), or when a true gradient is intractable and is replaced with a surrogate (e.g. in certain reinforcement learning applications or training networks with discrete variables). We propose Guided evolutionary Strategies, a method for optimally using surrogate gradient directions along with random search. We define a search distribution for evolutionary strategies that is elongated along a subspace spanned by the surrogate gradients. This allows us to estimate a descent direction which can then be passed to a first-order optimizer. We analytically and numerically characterize the tradeoffs that result from tuning how strongly the search distribution is stretched along the guiding subspace, and use this to derive a setting of the hyperparameters that works well across problems. Finally, we apply our method to example problems, demonstrating an improvement over both standard evolutionary strategies and first-order methods (that directly follow the surrogate gradient). We provide a demo of Guided ES at: ***/brain-research/guided-evolutionary-strategies. Copyright © 2018, The Authors. All rights reserved.

关键词： evolutionary algorithms

Evolving multimodal robot behavior via many stepping stones with the combinatorial multi-objective evolutionary algorithm

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Huizinga, Joost Clune, Jeff Evolving Artificial Intelligence Laboratory University of Wyoming LaramieWY82071 United States Uber AI Labs San FranciscoCA94104 United States

An important challenge in reinforcement learning, including evolutionary robotics, is to solve multimodal problems, where agents have to act in qualitatively different ways depending on the circumstances. Because multimodal problems are often too difficult to solve directly, it is helpful to take advantage of staging, where a difficult task is divided into simpler subtasks that can serve as stepping stones for solving the overall problem. Unfortunately, choosing an effective ordering for these subtasks is difficult, and a poor ordering can reduce the speed and performance of the learning process. Here, we provide a thorough introduction and investigation of the Combinatorial Multi-Objective evolutionary Algorithm (CMOEA), which avoids ordering subtasks by allowing all combinations of subtasks to be explored simultaneously. We compare CMOEA against two algorithms that can similarly optimize on multiple subtasks simultaneously: NSGA-II and Lexicase Selection. The algorithms are tested on a multimodal robotics problem with six subtasks as well as a maze navigation problem with a hundred subtasks. On these problems, CMOEA either outperforms or is competitive with the controls. Separately, we show that adding a linear combination over all objectives can improve the ability of NSGA-II to solve these multimodal problems. Lastly, we show that, in contrast to NSGA-II and Lexicase Selection, CMOEA can effectively leverage secondary objectives to achieve state-of the-art results on the robotics task. In general, our experiments suggest that CMOEA is a promising, state-of-the-art algorithm for solving multimodal problems. Copyright © 2018, The Authors. All rights reserved.

关键词： evolutionary algorithms