Differential evolution (DE) is a well-known type of evolutionary algorithms (EA). Similarly to other EA variants it can suffer from small populations and loose diversity too quickly. This paper presents a new approach to mitigate this issue: We propose to generate new candidate solutions by utilizing reversible linear transformation applied to a triplet of solutions from the population. In other words, the population is enlarged by using newly generated individuals without evaluating their fitness. We assess our methods on three problems: (i) benchmark function optimization, (ii) discovering parameter values of the gene repressilator system, (iii) learning neural networks. The empirical results indicate that the proposed approach outperforms vanilla DE and a version of DE with applying differential mutation three times on all testbeds.
Neural networks (NN) have been recently applied together with evolutionary algorithms (EAs) to solve dynamic optimization problems. The applied NN estimates the position of the next optimum based on the previous time best solutions. After detecting a change, the predicted solution can be employed to move the EA's population to a promising region of the solution space in order to accelerate convergence and improve accuracy in tracking the optimum. While previous works show improvement of the results, they neglect the overhead created by NN. In this work, we reflect the time spent on training NN in the optimization time and compare the results with a baseline EA. We explore if by considering the generated overhead, NN is still able to improve the results, and under which condition is able to do so. The main difficulties to train the NN are: 1) to get enough samples to generalize predictions for new data, and 2) to obtain reliable samples. As NN needs to collect data at each time step, if the time horizon is short, we will not be able to collect enough samples to train the NN. To alleviate this, we propose to consider more individuals on each change to speed up sample collection in shorter time steps. In environments with a high frequency of changes, the solutions produced by EA are likely to be far from the real optimum. Using unreliable train data for the NN will, in consequence, produce unreliable predictions. Also, as the time spent for NN stays fixed regardless of the frequency, a higher frequency of change will mean a higher produced overhead by the NN in proportion to the EA. In general, after considering the generated overhead, we conclude that NN is not suitable in environments with a high frequency of changes and/or short time horizons. However, it can be promising for the low frequency of changes, and especially for the environments that changes have a pattern.
Evolutionary algorithms, inspired by natural evolution, aim to optimize difficult objective functions without computing derivatives. Here we detail the relationship between population genetics and evolutionary optimization and formulate a new evolutionary algorithm. Optimization of a continuous objective function is analogous to searching for high fitness phenotypes on a fitness landscape. We summarize how natural selection moves a population along the non-euclidean gradient that is induced by the population on the fitness landscape (the natural gradient). Under normal approximations common in quantitative genetics, we show how selection is related to Newton's method in optimization. We find that intermediate selection is most informative of the fitness landscape. We describe the generation of new phenotypes and introduce an operator that recombines the whole population to generate variants that preserve normal statistics. Finally, we introduce a proof-of-principle algorithm that combines natural selection, our recombination operator, and an adaptive method to increase selection. Our algorithm is similar to covariance matrix adaptation and natural evolutionary strategies in optimization, and has similar performance. The algorithm is extremely simple in implementation with no matrix inversion or factorization, does not require storing a covariance matrix, and may form the basis of more general model-based optimization algorithms with natural gradient updates.
Water distribution system design is a challenging optimisation problem with a high number of search dimensions and constraints. In this way, Evolutionary Algorithms (EAs) have been widely applied to optimise WDS to minimise cost subject whilst meeting pressure constraints. This paper proposes a new hybrid evolutionary framework that consists of three distinct phases. The first phase applied CMA-ES, a robust adaptive meta-heuristic for continuous optimisation. This is followed by an upward-greedy search phase to remove pressure violations. Finally, a downward greedy search phase is used to reduce oversized pipes. To assess the effectiveness of the hybrid method, it was applied to five well-known WDSs case studies. The results reveal that the new framework outperforms CMA-ES by itself and other previously applied heuristics on most benchmarks in terms of both optimisation speed and network cost.
Designing evolutionary algorithms capable of uncovering highly evolvable representations is an open challenge; such evolvability is important because it accelerates evolution and enables fast adaptation to changing circumstances. This paper introduces evolvability ES, an evolutionary algorithm designed to explicitly and efficiently optimize for evolvability, i.e. the ability to further adapt. The insight is that it is possible to derive a novel objective in the spirit of natural evolution strategies that maximizes the diversity of behaviors exhibited when an individual is subject to random mutations, and that efficiently scales with computation. Experiments in 2-D and 3-D locomotion tasks highlight the potential of evolvability ES to generate solutions with tens of thousands of parameters that can quickly be adapted to solve different tasks and that can productively seed further evolution. We further highlight a connection between evolvability and a recent and popular gradient-based meta-learning algorithm called MAML; results show that evolvability ES can perform competitively with MAML and that it discovers solutions with distinct properties. The conclusion is that evolvability ES opens up novel research directions for studying and exploiting the potential of evolvable representations for deep neural networks.
Neural Architecture Search (NAS) has emerged as a promising technique for automatic neural network design. However, existing MCTS based NAS approaches often utilize manually designed action space, which is not directly related to the performance metric to be optimized (e.g., accuracy), leading to sample-inefficient explorations of architectures. To improve the sample efficiency, this paper proposes Latent Action Neural Architecture Search (LaNAS), which learns actions to recursively partition the search space into good or bad regions that contain networks with similar performance metrics. During the search phase, as different action sequences lead to regions with different performance, the search efficiency can be significantly improved by biasing towards the good regions. On three NAS tasks, empirical results demonstrate that LaNAS is at least an order more sample efficient than baseline methods including evolutionary algorithms, Bayesian optimizations, and random search. When applied in practice, both one-shot and regular LaNAS consistently outperform existing results. Particularly, LaNAS achieves 99.0% accuracy on CIFAR-10 and 80.8% top1 accuracy at 600 MFLOPS on ImageNet in only 800 samples, significantly outperforming AmoebaNet with 33x fewer samples. Our code is publicly available at https://github.com/facebookresearch/LaMCTS.
Evolution and learning are two of the fundamental mechanisms by which life adapts in order to survive and to transcend limitations. These biological phenomena inspired successful computational methods such as evolutionary algorithms and deep learning. Evolution relies on random mutations and on random genetic recombination. Here we show that learning to evolve, i.e. learning to mutate and recombine better than at random, improves the result of evolution in terms of fitness increase per generation and even in terms of attainable fitness. We use deep reinforcement learning to learn to dynamically adjust the strategy of evolutionary algorithms to varying circumstances. Our methods outperform classical evolutionary algorithms on combinatorial and continuous optimization problems.
Search-based procedural content generation uses stochastic global optimization algorithms to search for game content. However, standard tree search algorithms can be competitive with evolution on some optimization problems. We investigate the applicability of several tree search methods to level generation and compare them systematically with several optimization algorithms, including evolutionary algorithms. We compare them on three different game level generation problems: Binary, Zelda, and Sokoban. We introduce two new representations that can help tree search algorithms deal with the large branching factor of the generation problem. We find that in general, optimization algorithms clearly outperform tree search algorithms, but given the right problem representation certain tree search algorithms perform similarly to optimization algorithms, and in one particular problem, we see surprisingly strong results from MCTS.
Hard instances, which require a long time for a specific algorithm to solve, help (1) analyze the algorithm for accelerating it and (2) build a good benchmark for evaluating the performance of algorithms. There exist several efforts for automatic generation of hard instances. For example, evolutionary algorithms have been utilized to generate hard instances. However, they generate only finite number of hard instances. The merit of such methods is limited because it is difficult to extract meaningful patterns from small number of instances. We seek for a probabilistic generator of hard instances. Once the generative distribution of hard instances is obtained, we can sample a variety of hard instances to build a benchmark, and we can extract meaningful patterns of hard instances from sampled instances. The existing methods for modeling the hard instance distribution rely on parameters or rules that are found by domain experts; however, they are specific to the problem. Hence, it is challenging to model the distribution for general cases. In this paper, we focus on graph problems. We propose HiSampler, the hard instance sampler, to model the hard instance distribution of graph algorithms. HiSampler makes it possible to obtain the distribution of hard instances without hand-engineered features. To the best of our knowledge, this is the first method to learn the distribution of hard instances using machine learning. Through experiments, we demonstrate that our proposed method can generate instances that are a few to several orders of magnitude harder than the random-based approach in many settings. In particular, our method outperforms rule-based algorithms in the 3-coloring problem.
Reinforcement learning (RL) problems often feature deceptive local optima, and learning methods that optimize purely for reward signal often fail to learn strategies for overcoming them. Deep neuroevolution and novelty search have been proposed as effective alternatives to gradient-based methods for learning RL policies directly from pixels. In this paper, we introduce and evaluate the use of novelty search over agent action sequences by string edit metric distance as a means for promoting innovation. We also introduce a method for stagnation detection and population resampling inspired by recent developments in the RL community that uses the same mechanisms as novelty search to promote and develop innovative policies. Our methods extend a state-of-the-art method for deep neuroevolution using a simple-yet-effective genetic algorithm (GA) designed to efficiently learn deep RL policy network weights. Experiments using four games from the Atari 2600 benchmark were conducted. Results provide further evidence that GAs are competitive with gradient-based algorithms for deep RL. Results also demonstrate that novelty search over action sequences is an effective source of selection pressure that can be integrated into existing evolutionary algorithms for deep RL.
The one-fifth success rule is one of the best-known and most widely accepted techniques to control the parameters of evolutionary algorithms. While it is often applied in the literal sense, a common interpretation sees the one-fifth success rule as a family of success-based updated rules that are determined by an update strength $F$ and a success rate. We analyze in this work how the performance of the (1+1) Evolutionary Algorithm on LeadingOnes depends on these two hyper-parameters. Our main result shows that the best performance is obtained for small update strengths $F=1+o(1)$ and success rate $1/e$. We also prove that the running time obtained by this parameter setting is, apart from lower order terms, the same that is achieved with the best fitness-dependent mutation rate. We show similar results for the resampling variant of the (1+1) Evolutionary Algorithm, which enforces to flip at least one bit per iteration.
Very expensive problems are very common in practical system that one fitness evaluation costs several hours or even days. Surrogate assisted evolutionary algorithms (SAEAs) have been widely used to solve this crucial problem in the past decades. However, most studied SAEAs focus on solving problems with a budget of at least ten times of the dimension of problems which is unacceptable in many very expensive real-world problems. In this paper, we employ Voronoi diagram to boost the performance of SAEAs and propose a novel framework named Voronoi-based efficient surrogate assisted evolutionary algorithm (VESAEA) for very expensive problems, in which the optimization budget, in terms of fitness evaluations, is only 5 times of the problem's dimension. In the proposed framework, the Voronoi diagram divides the whole search space into several subspace and then the local search is operated in some potentially better subspace. Additionally, in order to trade off the exploration and exploitation, the framework involves a global search stage developed by combining leave-one-out cross-validation and radial basis function surrogate model. A performance selector is designed to switch the search dynamically and automatically between the global and local search stages. The empirical results on a variety of benchmark problems demonstrate that the proposed framework significantly outperforms several state-of-art algorithms with extremely limited fitness evaluations. Besides, the efficacy of Voronoi-diagram is furtherly analyzed, and the results show its potential to optimize very expensive problems.
Structured evolutionary algorithms have been investigated for some time. However, they have been under-explored specially in the field of multi-objective optimization. Despite their good results, the use of complex dynamics and structures make their understanding and adoption rate low. Here, we propose the general subpopulation framework that has the capability of integrating optimization algorithms without restrictions as well as aid the design of structured algorithms. The proposed framework is capable of generalizing most of the structured evolutionary algorithms, such as cellular algorithms, island models, spatial predator-prey and restricted mating based algorithms under its formalization. Moreover, we propose two algorithms based on the general subpopulation framework, demonstrating that with the simple addition of a number of single-objective differential evolution algorithms for each objective the results improve greatly, even when the combined algorithms behave poorly when evaluated alone at the tests. Most importantly, the comparison between the subpopulation algorithms and their related panmictic algorithms suggests that the competition between different strategies inside one population can have deleterious consequences for an algorithm and reveal a strong benefit of using the subpopulation framework. The code for SAN, the proposed multi-objective algorithm which has the current best results in the hardest benchmark, is available at the following https://github.com/zweifel/zweifel
In novel forms of the Social Internet of Things, any mobile user within communication range may help routing messages for another user in the network. The resulting message delivery rate depends both on the users' mobility patterns and the message load in the network. This new type of configuration, however, poses new challenges to security, amongst them, assessing the effect that a group of colluding malicious participants can have on the global message delivery rate in such a network is far from trivial. In this work, after modeling such a question as an optimization problem, we are able to find quite interesting results by coupling a network simulator with an evolutionary algorithm. The chosen algorithm is specifically designed to solve problems whose solutions can be decomposed into parts sharing the same structure. We demonstrate the effectiveness of the proposed approach on two medium-sized Delay-Tolerant Networks, realistically simulated in the urban contexts of two cities with very different route topology: Venice and San Francisco. In all experiments, our methodology produces attack patterns that greatly lower network performance with respect to previous studies on the subject, as the evolutionary core is able to exploit the specific weaknesses of each target configuration.
Many optimization techniques evaluate solutions consecutively, where the next candidate for evaluation is determined by the results of previous evaluations. For example, these include iterative methods, "black box" optimization algorithms, simulated annealing, evolutionary algorithms and tabu search, to name a few. When solving an optimization problem, these algorithms evaluate a large number of solutions, which raises the following question: Is it possible to learn something about the optimum using these solutions? In this paper, we define this "learning" question in terms of a logistic regression model and explore its predictive accuracy computationally. The proposed model uses a collection of solutions to predict the components of the optimal solutions. To illustrate the utility of such predictions, we embed the logistic regression model into the tabu search algorithm for job shop scheduling problem. The resulting framework is simple to implement, yet provides a significant boost to the performance of the standard tabu search.
Automatic neural architecture design has shown its potential in discovering powerful neural network architectures. Existing methods, no matter based on reinforcement learning or evolutionary algorithms (EA), conduct architecture search in a discrete space, which is highly inefficient. In this paper, we propose a simple and efficient method to automatic neural architecture design based on continuous optimization. We call this new approach neural architecture optimization (NAO). There are three key components in our proposed approach: (1) An encoder embeds/maps neural network architectures into a continuous space. (2) A predictor takes the continuous representation of a network as input and predicts its accuracy. (3) A decoder maps a continuous representation of a network back to its architecture. The performance predictor and the encoder enable us to perform gradient based optimization in the continuous space to find the embedding of a new architecture with potentially better accuracy. Such a better embedding is then decoded to a network by the decoder. Experiments show that the architecture discovered by our method is very competitive for image classification task on CIFAR-10 and language modeling task on PTB, outperforming or on par with the best results of previous architecture search methods with a significantly reduction of computational resources. Specifically we obtain 1.93% test set error rate for CIFAR-10 image classification task and 56.0 test set perplexity of PTB language modeling task. Furthermore, combined with the recent proposed weight sharing mechanism, we discover powerful architecture on CIFAR-10 (with error rate 2.93%) and on PTB (with test set perplexity 56.6), with very limited computational resources (less than 10 GPU hours) for both tasks.
The development, assessment, and comparison of randomized search algorithms heavily rely on benchmarking. Regarding the domain of constrained optimization, the number of currently available benchmark environments bears no relation to the number of distinct problem features. The present paper advances a proposal of a scalable linear constrained optimization problem that is suitable for benchmarking Evolutionary Algorithms. By comparing two recent EA variants, the linear benchmarking environment is demonstrated.
Multiobjective Evolutionary Algorithms based on Decomposition (MOEA/D) represent a widely used class of population-based metaheuristics for the solution of multicriteria optimization problems. We introduce the MOEADr package, which offers many of these variants as instantiations of a component-oriented framework. This approach contributes for easier reproducibility of existing MOEA/D variants from the literature, as well as for faster development and testing of new composite algorithms. The package offers an standardized, modular implementation of MOEA/D based on this framework, which was designed aiming at providing researchers and practitioners with a standard way to discuss and express MOEA/D variants. In this paper we introduce the design principles behind the MOEADr package, as well as its current components. Three case studies are provided to illustrate the main aspects of the package.
Fuzzing has become the de facto standard technique for finding software vulnerabilities. However, even state-of-the-art fuzzers are not very efficient at finding hard-to-trigger software bugs. Most popular fuzzers use evolutionary guidance to generate inputs that can trigger different bugs. Such evolutionary algorithms, while fast and simple to implement, often get stuck in fruitless sequences of random mutations. Gradient-guided optimization presents a promising alternative to evolutionary guidance. Gradient-guided techniques have been shown to significantly outperform evolutionary algorithms at solving high-dimensional structured optimization problems in domains like machine learning by efficiently utilizing gradients or higher-order derivatives of the underlying function. However, gradient-guided approaches are not directly applicable to fuzzing as real-world program behaviors contain many discontinuities, plateaus, and ridges where the gradient-based methods often get stuck. We observe that this problem can be addressed by creating a smooth surrogate function approximating the discrete branching behavior of target program. In this paper, we propose a novel program smoothing technique using surrogate neural network models that can incrementally learn smooth approximations of a complex, real-world program's branching behaviors. We further demonstrate that such neural network models can be used together with gradient-guided input generation schemes to significantly improve the fuzzing efficiency. Our extensive evaluations demonstrate that NEUZZ significantly outperforms 10 state-of-the-art graybox fuzzers on 10 real-world programs both at finding new bugs and achieving higher edge coverage. NEUZZ found 31 unknown bugs that other fuzzers failed to find in 10 real world programs and achieved 3X more edge coverage than all of the tested graybox fuzzers for 24 hours running.
Stochastic gradient descent is the most prevalent algorithm to train neural networks. However, other approaches such as evolutionary algorithms are also applicable to this task. Evolutionary algorithms bring unique trade-offs that are worth exploring, but computational demands have so far restricted exploration to small networks with few parameters. We implement an evolutionary algorithm that executes entirely on the GPU, which allows to efficiently batch-evaluate a whole population of networks. Within this framework, we explore the limited evaluation evolutionary algorithm for neural network training and find that its batch evaluation idea comes with a large accuracy trade-off. In further experiments, we explore crossover operators and find that unprincipled random uniform crossover performs extremely well. Finally, we train a network with 92k parameters on MNIST using an EA and achieve 97.6 % test accuracy compared to 98 % test accuracy on the same network trained with Adam. Code is available at https://github.com/jprellberg/gpuea.