Through the telecom lens: Are all training samples important?
This addresses the problem of high training demands and costs in telecommunications AI, offering a domain-specific incremental improvement for more sustainable systems.
The paper tackles the problem of inefficient AI training in telecommunications by questioning the assumption that all training samples are equally important, proposing a sample importance framework that selectively prioritizes impactful data to reduce computation without compromising accuracy. Experiments on three real-world telecom datasets show the method reserves performance while reducing data needs and computational overhead.
The rise of AI in telecommunications, from optimizing Radio Access Networks to managing user experience, has sharply increased data volumes and training demands. Telecom data is often noisy, high-dimensional, costly to store, process, and label. Despite Ai's critical role, standard workflows still assume all training samples contribute equally. On the other hand, next generation systems require AI models that are accurate, efficient, and sustainable.The paper questions the assumptions of equal importance by focusing on applying and analyzing the roles of individual samples in telecom training and assessing whether the proposed model optimizes computation and energy use. we perform sample-level gradient analysis across epochs to identify patterns of influence and redundancy in model learning. Based on this, we propose a sample importance framework thats electively prioritizes impactful data and reduces computation without compromising accuracy. Experiments on three real-world telecom datasets show that our method [reserves performance while reducing data needs and computational overhead while advancing the goals of sustainable AI in telecommunications.