A non-overlapped implantation MOSFET differential pair implementation of bidirectional weight update synapse for neuromorphic computing
| Autor: | Erik S. Jeng, Y. L. Chiang, J. Y. Chen, Hong-Xiu Chen, J. H. Chang |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
010302 applied physics
Quantitative Biology::Neurons and Cognition Artificial neural network Computer science Computer Science::Neural and Evolutionary Computation 020208 electrical & electronic engineering Transistor General Engineering 02 engineering and technology Chip Topology 01 natural sciences Iris flower data set law.invention Synaptic weight Neuromorphic engineering law 0103 physical sciences MOSFET 0202 electrical engineering electronic engineering information engineering Realization (systems) |
| Zdroj: | Microelectronics Journal. 90:306-314 |
| ISSN: | 0026-2692 |
| Popis: | A non-volatile memory (NVM) differential pair realization of the synaptic weight for an artificial neural network (ANN) circuit is investigated. Two non-overlapped implantation (NOI) nMOSFETs are proposed to form a NVM differential pair as an artificial synapse. The pair of NOI transistors are non-volatile analog memories and capable of storing positive or negative synaptic weight. In this study, a NOI differential pair based silicon neural chip of three neurons with 36 analog synapses in total is designed, simulated, fabricated, and verified by the chip-in-the-loop learning process. The classification performance of the neural chip when training and testing with the IRIS dataset is reported. This differential pair design not only overcomes the potential constraint in the weights of NOI but also exhibits better classification performances than that of single NOI-based ANN. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect