Curbing Negative Influences Online for Seamless Transfer Evolutionary Optimization

This paper draws motivation from the remarkable ability of humans to extract useful building-blocks of knowledge from past experiences and spontaneously reuse them for new and more challenging tasks. It is contended that successfully replicating such capabilities in computational solvers, particularly global black-box optimizers, can lead to significant performance enhancements over the current state-of-the-art. The main challenge to overcome is that in general black-box settings, no problem-specific data may be available prior to the onset of the search, thereby limiting the possibility of offline measurement of the synergy between problems. In light of the above, this paper introduces a novel evolutionary computation framework that enables online learning and exploitation of similarities across optimization problems, with the goal of achieving an algorithmic realization of the transfer optimization paradigm. One of the salient features of our proposal is that it accounts for latent similarities which while being less apparent on the surface, may be gradually revealed during the course of the evolutionary search. A theoretical analysis of our proposed framework is carried out, substantiating its positive influences on optimization performance. Furthermore, the practical efficacy of an instantiation of an adaptive transfer evolutionary algorithm is demonstrated on a series of numerical examples, spanning discrete, continuous, as well as singleand multi-objective optimization.

I. Introduction

In the global black-box optimization literature, efforts have seldom been made to automate the reuse of knowledge acquired from past problem-solving experiences. This limitation is primarily due to the lack of problem-specific data available prior to the onset of the search, which makes it difficult to ascertain (offline) the relationships across problems. In contrast, the idea of taking advantage of available data from various source tasks to improve the learning of a related target task has achieved significant success in the field of machine learning—under the label of transfer learning [1]–[5]. With this in mind, and under the observation that optimization problems of practical interest seldom exist in isolation [6], our goal in this paper is to achieve an algorithmic realization of the novel concept of transfer optimization [7]. Our contributions are expected to benefit real-world optimization settings of a time-sensitive nature, where ignoring prior experience implies the waste of a rich pool of knowledge that can otherwise be exploited to facilitate efficient re-exploration of possibly overlapping search spaces.