In this podcast episode, Kate discusses her groundbreaking research in using message passing neural networks for automatic chemical shift prediction in NMR spectra. She explains the intricate workings of MPNNs, highlighting their efficacy in handling molecular data. Despite challenges in neural network architecture and dataset size, Kate’s approach shows unprecedented precision, promising a significant impact on researchers and chemists using NMR spectroscopy. Additionally, she reflects on her academic journey, emphasizing her passion for computational statistics and machine learning, which led her to her current role at Mestrelab.

Ver todos los episodios