Start Small, Grow Big? Saving Multi-objective Function Evaluations
نویسندگان
چکیده
The influence of non-constant population sizes in evolutionary multi-objective optimization algorithms is investigated. In contrast to evolutionary single-objective optimization algorithms an increasing population size is considered beneficial when approaching the Pareto-front. Firstly, different deterministic schedules are tested, featuring different parameters like the initial population size. Secondly, a simple adaptation method is proposed. Considering all results, an increasing population size during an evolutionary multi-objective optimization algorithm run saves fitness function evaluations compared to a fixed population size. In particular, the results obtained with the adaptive method are most promising.
منابع مشابه
Multi-Objective Model for Fair Pricing of Electricity Using the Parameters from the Iran Electricity Market Big Data Analysis
Assessment of the electricity market shows that, electricity market data can be considered "big data". this data has been analyzed by both conventional and modern data mining methods. The predicted variables of supply and demand are considered to be the input of a defined multi-objective for predicting electricity price, which is the result of the defined model. This shows the advantage of appl...
متن کاملThermal-Economic Optimization of Shell and Tube Heat Exchanger by using a new Multi-Objective optimization method
Many studies are performed by researchers about Shell and Tube Heat Exchanger but the Multi-Objective Big Bang-Big Crunch algorithm (MOBBA) technique has never been used in such studies. This paper presents application of Thermal-Economic Multi-Objective Optimization of Shell and Tube Heat Exchanger Using MOBBA. For optimal design of a shell and tube heat exchanger, it was first thermally model...
متن کاملBig Models for Big Data using Multi objective averaged one dependence estimators
Even though, many researchers tried to explore the various possibilities on multi objective feature selection, still it is yet to be explored with best of its capabilities in data mining applications rather than going for developing new ones. In this paper, multi-objective evolutionary algorithm ENORA is used to select the features in a multi-class classification problem. The fusion of AnDE (av...
متن کاملBig Batch SGD: Automated Inference using Adaptive Batch Sizes
Classical stochastic gradient methods for optimization rely on noisy gradient approximations that become progressively less accurate as iterates approach a solution. The large noise and small signal in the resulting gradients makes it difficult to use them for adaptive stepsize selection and automatic stopping. We propose alternative “big batch” SGD schemes that adaptively grow the batch size o...
متن کاملMSPOT: Multi-Criteria Sequential Parmeter Optimization
Many relevant industrial optimization tasks feature more than just one quality criterion. State-of-the-art multi-criteria optimization algorithms require a relatively large number of function evaluations (usually more than 10) to approximate Pareto fronts. Due to high cost or time consumption this large amount of function evaluations is not always available. Therefore, it is obvious to combine ...
متن کامل