Graf Research and Virginia Tech are presenting a poster at the IEEE International Conference on Machine Learning for Communication and Networking (IEEE ICMLCN) 2024 in Stockholm, Sweden. The paper, titled "Exploring Explainable AI Techniques for Radio Frequency Machine Learning," is featured in Interactive Session 9 on Learning Communication Signal Processing on Wednesday, May 8th, from 1:30pm to 2:30pm.
The presentation highlights the challenges and solutions in interpreting complex deep learning models used in wireless radio frequency communications. The focus is on explainable artificial intelligence (XAI) techniques to demystify model decisions, specifically using attribution methods to assess the influence of inputs on outputs across different data modalities. This collaboration emphasizes the importance of transparency in AI systems, enhancing trust and understanding in advanced machine learning applications.
For additional details about the conference, visit IEEE ICMLCN 2024.
IEEE ICMLCN 2024 Interactive Session 9: Learning Communication Signal Processing
May 8, 2024, 1:30-2:30pm
Exploring Explainable AI Techniques for Radio Frequency Machine Learning
Stephen Adams (Virginia Tech, USA); Mia Taylor, Cody Crofford, Scott Harper, and Whitney Batchelor (Graf Research Corporation, USA); William C Headley (Virginia Tech, USA)
Abstract: Deep learning models are increasingly being used to solve complex wireless radio frequency communications problems. These state-of-the-art machine learning models have demonstrated superior performance over traditional methods when signal and environmental parameters are unknown a priori. However, due to the complexity of the architecture and the number of parameters, deep learning models are difficult to interpret. This opacity can lead to difficulties during testing and a lack of trust by the user. Explainable artificial intelligence (XAI) techniques can provide estimates for the impact an input has on the output of a model. In this study, we apply a wide range of common attribution techniques, a subset of XAI that focuses on estimating the contribution of each input to an output of a model, to simple wireless communications problems over two different data modalities (raw IQ and spectrogram images) and show how estimates of attributions could be used for test and evaluation.