Neuromorphic learning and communications
Neuromorphic learning leverages biologically inspired computational platforms that build on dynamic, sparse, event-driven signalling and processing. This talk will first present an overview of the state of the art by focusing on models and on the design of training algorithms. This will be done by distinguishing between deterministic and probabilistic models, as well as between solutions aimed at detecting spatial-only or spatio-temporal patterns. Then, two use cases involving communications will be outlined, namely neuromorphic federated learning and neuromorphic joint source-channel coding over wireless channels. The talk will also offer discussions on current limitations of the technology and on open problems.