Multifactorial Evolutionary Algorithm With Online Transfer Parameter Estimation: MFEA-II

Abstract

Humans rarely tackle every problem from scratch. Given this observation, the motivation for the present work is to improve optimization performance through adaptive knowledge transfer across related problems. The scope for spontaneous transfers under the simultaneous occurrence of multiple problems unveils the benefits of multitasking. Multitask optimization has recently demonstrated competence in solving multiple (related) optimization tasks concurrently. Notably, in the presence of underlying relationships between problems, the transfer of highquality solutions across them has shown to facilitate superior performance characteristics. However, in the absence of any prior knowledge about the inter-task synergies (as is often the case with general black-box optimization), the threat of predominantly negative transfer prevails. Susceptibility to negative inter-task interactions can impede the overall convergence behavior. To allay such fears, in this paper, we propose a novel evolutionary computation framework that enables online learning and exploitation of the similarities (and discrepancies) between distinct tasks in multitask settings, for an enhanced optimization process. Our proposal is based on principled theoretical arguments that seek to minimize the tendency of harmful interactions between tasks, based on a purely data-driven learning of relationships among them. The efficacy of our proposed method is validated experimentally on a series of synthetic benchmarks, as well as a practical study that provides insights into the behavior of the method in the face of several tasks occurring at once.

Rigorous Analysis of Multi-Factorial Evolutionary Algorithm as Multi-Population Evolution Model

“…In the literature, there exist a lot of works to apply MFEA to tackle real-world problems, such as complex supply chain network management [16], bi-level optimization problem [14], double-pole balancing problem [17], composites manufacturing problem [14,18], branch testing in software engineering [19], cloud computing service composition problem [20], pollution-routing problem [21], operational indices optimization of beneficiation process [22], and time series prediction problem [23].…”

“…The most widely utilized one is probably genetic mechanisms, namely crossover and mutation. Specifically, several typical genetic strategies include SBX [6,17], ordered crossover [25], one-point crossover [26], guided differential evolutionary crossover [27], Gaussian mutation [6], swap mutation [25], polynomial mutation [17,28], and swap-change mutation [29]. The other three EAs, differential evolution (DE) [13,[30][31][32], particle swarm optimization (PSO) [31,[33][34][35], and genetic programming (GP) [23], are also utilized as fundamental algorithm for MTO paradigms.…”

“…where is a constant parameter, and N(0, 1) is a Gaussian noise with zero mean and unit variance. It is very recently reported that an online rmp estimation technique was proposed in order to theoretically minimize the negative interactions between distinct optimization tasks [17]. Specifically, the extent of transfer parameter matrix is learned and adapted online based on the optimal blending of probabilistic models in a purely data-driven manner.…”

A Fireworks Algorithm Based on Transfer Spark for Evolutionary Multitasking

“…Currently, the research on EMT can approximately be summarized into three categories, the practical application of EMT (Sagarna and Ong, 2016;Yuan et al, 2016;Zhou et al, 2016;Cheng et al, 2017;Binh et al, 2018;Thanh et al, 2018;Lian et al, 2019;Wang et al, 2019) and the improved algorithm based on the MFEA framework (Bali et al, 2017;Feng et al, 2017;Wen and Ting, 2017;Joy et al, 2018;Tuan et al, 2018;Zhong et al, 2018;Binh et al, 2019;Liang et al, 2019;Yin et al, 2019;Yu et al, 2019;Zheng et al, 2019;Zhou et al, 2019) and the perfection of EMT theory (Gupta et al, 2016a;Hashimoto et al, 2018;Liu et al, 2018;Zhou et al, 2018;Bali et al, 2019;Chen et al, 2019;Feng et al, 2019;Huang et al, 2019;Shang et al, 2019;Song et al, 2019;Tang et al, 2019). From the above studies, a consensus can be summarized that efficiently utilizing the inter-task related information is the key to improve overall search efficiency in EMT.…”

A Preliminary Study of Knowledge Transfer in Multi-Classification Using Gene Expression Programming

“…In fact, intuitively, a classification rule trained by binary classifiers of one class can hopefully be utilized by another class as a rule component that can to some extent boost its own binary classification performance through recognizing the pattern of negative samples. According to the consideration above, this paper takes into account the Evolutionary Multitasking paradigm (Gupta et al, 2015(Gupta et al, , 2017Ong and Gupta, 2016;Bali et al, 2019) to facilitate the multi-classification avoiding output collision of binary classifiers by enhancing the knowledge transfer among multiple binary classifiers. Equipped with the capability of latent genetic transfer, Evolutionary Multitasking can resolve many optimization problems simultaneously by enabling the knowledge transfer among different problems through the unified chromosome representation.…”