Machine learning to decipher immune recognition (Doctor AI)

The adaptive immune system records all past and ongoing battles with disease and infection in the form of immune memory, stored in the form of DNA of immune receptors of adaptive immune cells. However, deciphering these signals is a grand challenge of immunology, requiring sophisticated machine learning.

While a myriad of approaches has been proposed, they do not exploit all available domain knowledge, and lack interoperability, reproducibility and comparative benchmarking. We are developing novel approaches for incorporating domain knowledge into deep learning models, and are also defining improved benchmarks and a machine learning software platform called immuneML to unify ML methodology in the field.

Tags: Bayesian machine learning, Causality, Data fusion/integration, Deep learning/neural networks, Explainable AI, Generative machine learning, Graphical modeling, High dimensional inference, Language models, Multiple instant learning, Regularized learning, Software framework development, Transfer learning, Unsupervised learning, Machine Learning/Artificial Intelligence

Published July 6, 2023 2:45 PM - Last modified Oct. 23, 2023 11:48 AM