A fully flexible circuit implementation of clique-based neural networks in 65-nm CMOS
- Resource Type
- Conference
- Authors
- Larras, Benoit; Chollet, Paul; Lahuec, Cyril; Seguin, Fabrice; Arzel, Matthieu
- Source
- 2018 IEEE International Symposium on Circuits and Systems (ISCAS) Circuits and Systems (ISCAS), 2018 IEEE International Symposium on. :1-4 May, 2018
- Subject
- Components, Circuits, Devices and Systems
Power, Energy and Industry Applications
Signal Processing and Analysis
Neurons
Synapses
Neural networks
Electrocardiography
Image color analysis
Complexity theory
Computer architecture
- Language
- ISSN
- 2379-447X
Clique-based neural networks implement low-complexity functions working with a reduced connectivity between neurons. Thus, they address very specific applications operating with a very low energy budget. This paper proposes a flexible and iterative neural architecture able to implement multiple types of clique-based neural networks of up to 3968 neurons. The circuit has been integrated in a ST 65-nm CMOS ASIC and validated in the context of ECG classification. The network core reacts in 83ns to a stimulation and occupies a 0.21mm 2 silicon area.