In multi-task constrained multi-objective optimization, existing studies have not fully considered the similarity between transfer tasks. To address this issue, we propose a novel multi-task constrained multi-objectiv...
详细信息
The image synthesis technique is relatively well established which can generate facial images that are indistinguishable even by human beings. However, all of these approaches uses gradients to condition the output, r...
详细信息
The image synthesis technique is relatively well established which can generate facial images that are indistinguishable even by human beings. However, all of these approaches uses gradients to condition the output, resulting in the outputting the same image with the same input. Also, they can only generate images with basic expression or mimic an expression instead of generating compound expression. In real life, however, human expressions are of great diversity and complexity. In this paper, we propose an evo-lutionary algorithm (EA) assisted GAN, named EvoGAN, to generate various compound expressions with any accurate target compound expression. EvoGAN uses an EA to search target results in the data distri-bution learned by GAN. Specifically, we use the Facial Action Coding System (FACS) as the encoding of an EA and use a pre-trained GAN to generate human facial images, and then use a pre-trained classifier to recognize the expression composition of the synthesized images as the fitness function to guide the search of the EA. Combined random searching algorithm, various images with the target expression can be easily sythesized. Quantitative and Qualitative results are presented on several compound expres-sions, and the experimental results demonstrate the feasibility and the potential of EvoGAN. The source code is available at https://***/ECNU-Cross-Innovation-Lab/EvoGAN. (c) 2021 Elsevier B.V. All rights reserved.
The effectiveness of the machine learning methods for real-world tasks depends on the proper structure of the modeling pipeline. The proposed approach is aimed to automate the design of composite machine learning pipe...
详细信息
The effectiveness of the machine learning methods for real-world tasks depends on the proper structure of the modeling pipeline. The proposed approach is aimed to automate the design of composite machine learning pipelines, which is equivalent to computation workflows that consist of models and data operations. The approach combines key ideas of both automated machine learning and workflow management systems. It designs the pipelines with a customizable graph-based structure, analyzes the obtained results, and reproduces them. The evolutionary approach is used for the flexible identification of pipeline structure. The additional algorithms for sensitivity analysis, atomization, and hyperparameter tuning are implemented to improve the effectiveness of the approach. Also, the software implementation on this approach is presented as an open-source framework. The set of experiments is conducted for the different datasets and tasks (classification, regression, time series forecasting). The obtained results confirm the correctness and effectiveness of the proposed approach in the comparison with the state-of-the-art competitors and baseline solutions. (C) 2021 Elsevier B.V. All rights reserved.
Demand-Side Management (DSM) refers to programs that aim to control the energy consumption at the customer side of the meter. Different techniques have been proposed to achieve this. Perhaps the most popular technique...
详细信息
ISBN:
(纸本)9781450324694
Demand-Side Management (DSM) refers to programs that aim to control the energy consumption at the customer side of the meter. Different techniques have been proposed to achieve this. Perhaps the most popular techniques are those based on smart pricing (e.g., critical-peak pricing, real-time pricing). The idea, in a nutshell, is to encourage end users to shift their load consumption based on the price at a particular time (e.g., the higher the price, the less number of electric appliances are expected to be turned on). Motivated by these techniques (e.g., a strong positive correlation between the number of appliances being used and the electricity cost), we propose the use of an stochastic evolutionary-based optimisation technique, evolutionary algorithms, to automatically generate optimal, or nearly optimal, solutions that represent schedules to charge a number of electric vehicles (EVs) with two goals: (a) that each EV is as fully charged as possible at time of departure, and (b) to avoid charging them at the same time, whenever possible (e.g., load reduction at the transformer level). Instead of using a price signal to shift load consumption, we achieve this by considering what all the EVs might do at a particular time, rather than considering an interaction between an utility company and its user, as normally adopted in DSM programs. We argue that exploiting the interaction of these EVs is crucial at achieving excellent results because it carries the notion of smart pricing (e.g., balance energy usage), which is highly popular in DSM systems. Thus, the main contribution of this work is the notion of load shifting, borrowed from smart pricing methods, implemented in an evolutionary-based algorithm to automatically generate optimal solutions. To test our proposed approach, we used a dynamic scenario, where the state of charge of each EV is different for every day of our 28 days testing period. The results obtained by our proposed approach are highly encouraging in
The differential evolution algorithm has rich, successful experience in parameter settings. How to reasonably control strategies and parameters and effectively utilize feedback information from individuals in the popu...
详细信息
Nowadays, air pollution forecasting modeling is vital to achieve an increase in air quality, allowing an improvement of ecosystems and human health. It is important to consider the spatial characteristics of the data,...
详细信息
Nowadays, air pollution forecasting modeling is vital to achieve an increase in air quality, allowing an improvement of ecosystems and human health. It is important to consider the spatial characteristics of the data, as they allow us to infer predictions in those areas for which no information is available. In the current literature, there are a large number of proposals for spatio-temporal air pollution forecasting. In this paper we propose a novel spatio-temporal approach based on multi-objective evolutionary algorithms for the identification of multiple non-dominated linear regression models and their combination in an ensemble learning model for air pollution forecasting. The ability of multi-objective evolutionary algorithms to find a Pareto front of solutions is used to build multiple forecast models geographically distributed in the area of interest. The proposed method has been applied for one-week NO2 prediction in southeastern Spain and has obtained promising results in statistical comparison with other approaches such as the union of datasets or the interpolation of the predictions for each monitoring station. The validity of the proposed spatio-temporal approach is thus demonstrated, opening up a new field in air pollution engineering. (C) 2022 The Author(s). Published by Elsevier B.V.
Understanding how evolutionary agents behave in complex environments is a challenging problem. Agents can be faced with complex fitness landscapes derived from multi-stage tasks, interaction with others, and limited e...
详细信息
Understanding how evolutionary agents behave in complex environments is a challenging problem. Agents can be faced with complex fitness landscapes derived from multi-stage tasks, interaction with others, and limited environmental feedback. Agents that evolve to overcome these can sometimes access greater fitness, as a result of factors such as cooperation and tool use. However, it is often difficult to explain why evolutionary agents behave in certain ways, and what specific elements of the environment or task may influence the ability of evolution to find goal-achieving behaviours;even seemingly simple environments or tasks may contain features that affect agent evolution in unexpected ways. We explore principled simplification of evolutionary agent-based models, as a possible route to aiding their explainability. Using the River Crossing Task (RCT) as a case study, we draw on analysis in the Minimal River Crossing (RC-) Task testbed, which was designed to simplify the original task while keeping its key features. Using this method, we present new analysis concerning when agents evolve to successfully complete the RCT. We demonstrate that the RC- environment can be used to understand the effect that a cost to movement has on agent evolution, and that these findings can be generalised back to the original RCT. Then, we present new insight into the use of principled simplification in understanding evolutionary agents. We find evidence that behaviour dependent on features that survive simplification, such as problem structure, are amenable to prediction;while predicting behaviour dependent on features that are typically reduced in simplification, such as scale, can be invalid.
Finding optimal parameters for detailed neuronal models is a ubiquitous challenge in neuroscientific research. In recent years, manual model tuning has been gradually replaced by automated parameter search using a var...
详细信息
Finding optimal parameters for detailed neuronal models is a ubiquitous challenge in neuroscientific research. In recent years, manual model tuning has been gradually replaced by automated parameter search using a variety of different tools and methods. However, using most of these software tools and choosing the most appropriate algorithm for a given optimization task require substantial technical expertise, which prevents the majority of researchers from using these methods effectively. To address these issues, we developed a generic platform (called Neuroptimus) that allows users to set up neural parameter optimization tasks via a graphical interface, and to solve these tasks using a wide selection of state-of-the-art parameter search methods implemented by five different Python packages. Neuroptimus also offers several features to support more advanced usage, including the ability to run most algorithms in parallel, which allows it to take advantage of high-performance computing architectures. We used the common interface provided by Neuroptimus to conduct a detailed comparison of more than twenty different algorithms (and implementations) on six distinct benchmarks that represent typical scenarios in neuronal parameter search. We quantified the performance of the algorithms in terms of the best solutions found and in terms of convergence speed. We identified several algorithms, including covariance matrix adaptation evolution strategy and particle swarm optimization, that consistently, without any fine-tuning, found good solutions in all of our use cases. By contrast, some other algorithms including all local search methods provided good solutions only for the simplest use cases, and failed completely on more complex problems. We also demonstrate the versatility of Neuroptimus by applying it to an additional use case that involves tuning the parameters of a subcellular model of biochemical pathways. Finally, we created an online database that allows uploading,
Multi objective optimization evolutionary algorithms (MOEAs) play a crucial role in addressing multi-objective optimization problems (MOPs) in the field of artificial intelligence. However, MOEAs often struggle to sim...
详细信息
暂无评论