Domain wall and Magnetic Tunnel Junction Hybrid for on-chip Learning in UNet architecture
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910522100678656 |
|---|---|
| author | Vadde, Venkatesh Muralidharan, Bhaskaran Sharma, Abhishek |
| author_facet | Vadde, Venkatesh Muralidharan, Bhaskaran Sharma, Abhishek |
| contents | We present spintronic devices based hardware implementation of UNet for segmentation tasks. Our approach involves designing hardware for convolution, deconvolution, rectified activation function (ReLU), and max pooling layers of the UNet architecture. We designed the convolution and deconvolution layers of the network using the synaptic behavior of the domain wall MTJ. We also construct the ReLU and max pooling functions of the network utilizing the spin hall driven orthogonal current injected MTJ. To incorporate the diverse physics of spin-transport, magnetization dynamics, and CMOS elements in our UNet design, we employ a hybrid simulation setup that couples micromagnetic simulation, non-equilibrium Green's function, SPICE simulation along with network implementation. We evaluate our UNet design on the CamVid dataset and achieve segmentation accuracies of 83.71$\%$ on test data, on par with the software implementation with 821mJ of energy consumption for on-chip training over 150 epochs. We further demonstrate nearly one order $(10\times)$ improvement in the energy requirement of the network using unstable ferromagnet ($Δ$=4.58) over the stable ferromagnet ($Δ$=45) based ReLU and max pooling functions while maintaining the similar accuracy. The hybrid architecture comprising domain wall MTJ and unstable FM-based MTJ leads to an on-chip energy consumption of 85.79mJ during training, with a testing energy cost of 1.55 $μJ$. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_02863 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Domain wall and Magnetic Tunnel Junction Hybrid for on-chip Learning in UNet architecture Vadde, Venkatesh Muralidharan, Bhaskaran Sharma, Abhishek Emerging Technologies Image and Video Processing Applied Physics We present spintronic devices based hardware implementation of UNet for segmentation tasks. Our approach involves designing hardware for convolution, deconvolution, rectified activation function (ReLU), and max pooling layers of the UNet architecture. We designed the convolution and deconvolution layers of the network using the synaptic behavior of the domain wall MTJ. We also construct the ReLU and max pooling functions of the network utilizing the spin hall driven orthogonal current injected MTJ. To incorporate the diverse physics of spin-transport, magnetization dynamics, and CMOS elements in our UNet design, we employ a hybrid simulation setup that couples micromagnetic simulation, non-equilibrium Green's function, SPICE simulation along with network implementation. We evaluate our UNet design on the CamVid dataset and achieve segmentation accuracies of 83.71$\%$ on test data, on par with the software implementation with 821mJ of energy consumption for on-chip training over 150 epochs. We further demonstrate nearly one order $(10\times)$ improvement in the energy requirement of the network using unstable ferromagnet ($Δ$=4.58) over the stable ferromagnet ($Δ$=45) based ReLU and max pooling functions while maintaining the similar accuracy. The hybrid architecture comprising domain wall MTJ and unstable FM-based MTJ leads to an on-chip energy consumption of 85.79mJ during training, with a testing energy cost of 1.55 $μJ$. |
| title | Domain wall and Magnetic Tunnel Junction Hybrid for on-chip Learning in UNet architecture |
| topic | Emerging Technologies Image and Video Processing Applied Physics |
| url | https://arxiv.org/abs/2403.02863 |