Learning to Optimize: Unrolling-based Approach for 6G Wireless Networks

  • Nguyen, The Vi; 
  • Nguyen, Thi My Tuyen; 
  • Song, Chihyun; 
  • Cho, Sungrae
Citations

SCOPUS

0

초록

Deep neural networks (DNNs) have shown strong capability in wireless optimization, but often lack interpretability and theoretical guarantees. Traditional iterative algorithms are interpretable but can be inefficient for large-scale systems. Algorithm unrolling combines their strengths by representing each iteration as a neural network layer, enabling fast learning, scalability, and interpretability. This paper reviews the principles of algorithm unrolling and its integration into problems in intelligent reflecting surface (IRS)-assisted networks, demonstrating its potential to achieve high performance with reduced computational complexity.

키워드

6G networks; Algorithm unrolling; Intelligent reflecting surface (IRS); Machine learning
제목
Learning to Optimize: Unrolling-based Approach for 6G Wireless Networks
저자
Nguyen, The Vi; Nguyen, Thi My Tuyen; Song, Chihyun; Cho, Sungrae
DOI
10.1109/ICTC66702.2025.11388458
발행일
2025
유형
Conference Paper
저널명
International Conference on ICT Convergence
페이지
681 ~ 683