Training Deep Convolutional Neural Networks with Resistive Cross-Point Devices
In a previous work we have detailed the requirements for obtaining maximal deep learning performance benefit by implementing fully connected deep neural networks (DNN) in the form of arrays of resistive devices. Here we extend the concept of Resistive Processing Unit (RPU) devices to convolutional n...
Autori principali: | Tayfun Gokmen, Murat Onen, Wilfried Haensch |
---|---|
Natura: | Articolo |
Lingua: | English |
Pubblicazione: |
Frontiers Media S.A.
2017-10-01
|
Serie: | Frontiers in Neuroscience |
Soggetti: | |
Accesso online: | http://journal.frontiersin.org/article/10.3389/fnins.2017.00538/full |
Documenti analoghi
Documenti analoghi
-
Training LSTM Networks With Resistive Cross-Point Devices
di: Tayfun Gokmen, et al.
Pubblicazione: (2018-10-01) -
MR-PIPA: An Integrated Multilevel RRAM (HfO<sub><italic>x</italic></sub>)-Based Processing-In-Pixel Accelerator
di: Minhaz Abedin, et al.
Pubblicazione: (2022-01-01) -
Reliable and Energy-Efficient Diabetic Retinopathy Screening Using Memristor-Based Neural Networks
di: Sumit Diware, et al.
Pubblicazione: (2024-01-01) -
Design implementations of ternary logic systems: A critical review
di: Furqan Zahoor, et al.
Pubblicazione: (2024-09-01) -
Accurate Inference With Inaccurate RRAM Devices: A Joint Algorithm-Design Solution
di: Gouranga Charan, et al.
Pubblicazione: (2020-01-01)