Siamese Meta-Learning and Algorithm Selection with 'Algorithm-Performance Personas' [Proposal]

TL;DR

This paper introduces a Siamese neural network architecture for algorithm selection that emphasizes 'alike performing' instances, utilizing a novel performance metric and the concept of 'Algorithm Performance Personas' to improve training sample selection.

Contribution

It proposes a new Siamese network approach focusing on instance similarity, along with a novel performance metric and the concept of 'Algorithm Performance Personas' for better meta-learning.

Findings

The proposed metric outperforms standard absolute error metrics.

The concept of 'alike performing' instances aids in training sample selection.

Initial evidence shows improved suitability for training sample selection.

Abstract

Automated per-instance algorithm selection often outperforms single learners. Key to algorithm selection via meta-learning is often the (meta) features, which sometimes though do not provide enough information to train a meta-learner effectively. We propose a Siamese Neural Network architecture for automated algorithm selection that focuses more on 'alike performing' instances than meta-features. Our work includes a novel performance metric and method for selecting training samples. We introduce further the concept of 'Algorithm Performance Personas' that describe instances for which the single algorithms perform alike. The concept of 'alike performing algorithms' as ground truth for selecting training samples is novel and provides a huge potential as we believe. In this proposal, we outline our ideas in detail and provide the first evidence that our proposed metric is better suitable for training sample selection that standard performance metrics such as absolute errors.