CSIC - Sevilla, Spain

Digital Twin of the Neuromorphic Device

Aims of this project are: (i) to develop and demonstrate electrical measurements of the memristive devices and to compare them with SPICE behavioural models; (ii) derive guidelines for device optimization for the learning and computation of artificial intelligence tasks; (iii) Demonstrate by simulation the learning of the complete set of digital logic operations; (iv) Development of abstract models for Python code in spiking neural networks simulation frameworks; (v) Benchmark the designed neural networks in classification of standards benchmarks such as the MNIST or the Iris database; (vi) Study of delay learning for recognition of spatio-temporal data.

 

Prof. Teresa Serrano-Gotarredona

Principal Investigator

Planned Secondments