Energy and Quality of Surrogate-Assisted Search Algorithms: a First Analysis

2025-08-11

AI Summary

Addressing the lack of integrated evaluation of energy efficiency and solution accuracy in surrogate-assisted metaheuristic algorithms, this paper proposes the first comprehensive assessment framework that jointly models CPU/memory energy consumption and surrogate model prediction accuracy. Using Particle Swarm Optimization (PSO) as the baseline optimizer and a pre-trained–fine-tuned neural network as the surrogate, we conduct multi-configuration comparative experiments to quantify both computational energy profiles and search quality under varying surrogate strategies. Results reveal a pronounced trade-off between surrogate accuracy and system-level energy efficiency, demonstrating that higher prediction accuracy does not necessarily improve overall energy efficiency. We introduce the “energy-efficiency–accuracy dual-dimension evaluation” paradigm, providing theoretical foundations and empirical evidence for the design, selection, and sustainable deployment of green intelligent optimization algorithms.

Technology Category

• Search and Optimization: Metareasoning and Metaheuristics
• Machine Learning: Efficient ML / Green AI
• Constraint Satisfaction and Optimization: Constraint Optimization

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success

Abstract

Solving complex real problems often demands advanced algorithms, and then continuous improvements in the internal operations of a search technique are needed. Hybrid algorithms, parallel techniques, theoretical advances, and much more are needed to transform a general search algorithm into an efficient, useful one in practice. In this paper, we study how surrogates are helping metaheuristics from an important and understudied point of view: their energy profile. Even if surrogates are a great idea for substituting a time-demanding complex fitness function, the energy profile, general efficiency, and accuracy of the resulting surrogate-assisted metaheuristic still need considerable research. In this work, we make a first step in analyzing particle swarm optimization in different versions (including pre-trained and retrained neural networks as surrogates) for its energy profile (for both processor and memory), plus a further study on the surrogate accuracy to properly drive the search towards an acceptable solution. Our conclusions shed new light on this topic and could be understood as the first step towards a methodology for assessing surrogate-assisted algorithms not only accounting for time or numerical efficiency but also for energy and surrogate accuracy for a better, more holistic characterization of optimization and learning techniques.

Problem

• Analyzing energy profile of surrogate-assisted metaheuristics
• Evaluating efficiency and accuracy of surrogate models
• Assessing PSO variants with neural network surrogates

Innovation

• Analyzes energy profile of surrogate-assisted metaheuristics
• Uses pre-trained and retrained neural networks as surrogates
• Assesses surrogate accuracy for holistic algorithm evaluation