Calibrating AI Models for Few-Shot Demodulation VIA Conformal Prediction
Calibrating AI Models for Few-Shot Demodulation VIA Conformal Prediction
复制标题
DOI：
10.1109/icassp49357.2023.10096780
复制
发表时间：
2022-10
期刊：
影响因子：
--
通讯作者：
K. Cohen;Sangwoo Park;O. Simeone;S. Shamai
中科院分区：
文献类型：
--
作者：
K. Cohen;Sangwoo Park;O. Simeone;S. Shamai
人工智能（AI）工具可以用于解决通信系统设计中的模型缺陷。然而，传统的基于学习的人工智能算法产生的决策校准较差，无法量化其输出的不确定性。虽然贝叶斯学习可以通过捕获由有限的数据可用性引起的认知不确定性来增强校准，但正式校准保证仅在关于地面真实、未知数据生成机制的强假设下才有效。我们建议利用共形预测框架来获得数据驱动的集合预测，其校准属性与数据分布无关。具体来说，我们研究了基带解调器的设计中存在的难以建模的非线性，如硬件缺陷，并提出了基于共形预测的解调器。数值结果证实了所提出的解调器的理论有效性，并带来洞察其平均预测集大小的效率。
Artificial Intelligent (AI) tools can be useful to address model deficits in the design of communication systems. However, conventional learning-based AI algorithms yield poorly calibrated decisions, unabling to quantify their outputs uncertainty. While Bayesian learning can enhance calibration by capturing epistemic uncertainty caused by limited data availability, formal calibration guarantees only hold under strong assumptions about the ground-truth, unknown, data generation mechanism. We propose to leverage the conformal prediction framework to obtain data-driven set predictions whose calibration properties hold irrespective of the data distribution. Specifically, we investigate the design of baseband demodulators in the presence of hard-to-model nonlinearities such as hardware imperfections, and propose set-based demodulators based on conformal prediction. Numerical results confirm the theoretical validity of the proposed demodulators, and bring insights into their average prediction set size efficiency.