Presentation Information

4:15 PM - 4:30 PM JST(7:15 AM - 7:30 AM UTC)Candidate for Academic Incentive Award

[B-15-23]A Study on Channel-Adaptive Feature-Budget Optimization for Semantic Communication

◎Zhiyi Zhu¹, Eiji Takimoto², Patrick Finnerty¹, Chikara Ohta¹ (1. Kobe University, 2. Nara Women's University)

Keywords:

Semantic communication,Optimization

Semantic communication reduces transmission cost by sending task-relevant features, but transmitted feature budgets are usually fixed and ignore channel quality. This study optimizes transmitted features while keeping task accuracy close to a fixed feature.

Back to Session information