Hyper-parameter Optimization for Wireless Network Traffic Prediction Models with A Novel Meta-Learning Framework

2024-09-22

arXiv.org

AI Summary

To address the low efficiency of hyperparameter optimization and poor cross-task generalization in wireless network traffic forecasting, this paper proposes a meta-learning-driven automated hyperparameter optimization framework. The method employs an attention-based deep neural network as the base learner; task-specific meta-features guide a K-nearest neighbors (KNN) search for candidate optimization strategies, while an enhanced genetic algorithm—incorporating intelligent chromosome selection—dynamically generates optimal hyperparameters. Its core innovation lies in the first integration of meta-feature-guided KNN retrieval with a search-efficient genetic algorithm, enabling synergistic cross-task knowledge transfer and task-specific hyperparameter tuning. Evaluated on real-world wireless traffic datasets, the framework reduces the base learner’s mean absolute error (MAE) by 18.7% on average and accelerates hyperparameter search by 5.3×, significantly improving both predictive accuracy and generalization capability.

Technology Category

• Search and Optimization: Metareasoning and Metaheuristics
• Machine Learning: Auto ML and Hyperparameter Tuning
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

In this paper, we propose a novel meta-learning based hyper-parameter optimization framework for wireless network traffic prediction models. An attention-based deep neural network (ADNN) is adopted as the prediction model, i.e., base-learner, for each wireless network traffic prediction task, namely base-task, and a meta-learner is employed to automatically generate the optimal hyper-parameters for a given base-learner according to the corresponding base-task's intrinsic characteristics or properties, i.e., meta-features. Based on our observation from real-world traffic records that base-tasks possessing similar meta-features tend to favour similar hyper-parameters for their base-learners, the meta-learner exploits a K-nearest neighbor (KNN) learning method to obtain a set of candidate hyper-parameter selection strategies for a new base-learner, which are then utilized by an advanced genetic algorithm with intelligent chromosome screening to finally acquire the best hyper-parameter selection strategy. Extensive experiments demonstrate that base-learners in the proposed framework have high potential prediction ability for wireless network traffic prediction task, and the meta-learner can enormously elevate the base-learners' performance by providing them the optimal hyper-parameters.

Problem

• Optimizing hyper-parameters for wireless traffic prediction models
• Accelerating hyper-parameter selection using meta-learning and KNN-GA-GRN
• Improving prediction accuracy and efficiency over traditional methods

Innovation

• Meta-learning framework optimizes hyper-parameters efficiently
• Attention-based DNN and KNN-GA-GRN integration
• Accelerates hyper-parameter validation with GRN screening