Towards Exploratory Landscape Analysis for Large-scale Optimization: A Dimensionality Reduction Framework
This work addresses scalability issues in ELA for researchers and practitioners in optimization, though it is incremental as it builds on existing methods by improving efficiency.
The paper tackled the problem of scaling exploratory landscape analysis (ELA) to large-scale optimization by analyzing computational costs and proposing a dimensionality reduction framework, which drastically reduced computation time for key feature classes and made cell-mapping features scalable, showing benefits for predicting properties of 24 large-scale BBOB functions.
Although exploratory landscape analysis (ELA) has shown its effectiveness in various applications, most previous studies focused only on low- and moderate-dimensional problems. Thus, little is known about the scalability of the ELA approach for large-scale optimization. In this context, first, this paper analyzes the computational cost of features in the flacco package. Our results reveal that two important feature classes (ela_level and ela_meta) cannot be applied to large-scale optimization due to their high computational cost. To improve the scalability of the ELA approach, this paper proposes a dimensionality reduction framework that computes features in a reduced lower-dimensional space than the original solution space. We demonstrate that the proposed framework can drastically reduce the computation time of ela_level and ela_meta for large dimensions. In addition, the proposed framework can make the cell-mapping feature classes scalable for large-scale optimization. Our results also show that features computed by the proposed framework are beneficial for predicting the high-level properties of the 24 large-scale BBOB functions.