Next-generation adaptive electronics for neuromorphic engineering
1 September 2017
Exploiting the potential of RRAM technology for extremely power-efficient neuromorphic systems, specifically Hardware Accelerators and Physical Neural Networks for ML applications
听
Funder
Amount 拢 500 000
Project website听
Description
This project takes a fundamentally different approach to create a new neuromorphic technology. Resistive RAM (RRAM) technology, a subclass of memristive systems, is based on simple two terminals (metal-oxide-metal) nanodevices whose resistance can repeatedly be varied, with low operational energy & very high levels of integration. Apart from memory, memristors also enable computing. This is a crucial step in replacing a von-Neumann鈥檚 back-and-forth bottleneck by architectures that directly implement neuromorphic functions 鈥 most importantly synapse-like plasticity, and neuron-like integration & spiking.
Neuromorphic memristive systems promise to yield vast improvements in power efficiency, therefore, enhancing artificial intelligence in mobile and embedded systems. This can be leveraged in the 鈥淚ntelligence of things鈥 era with ML algorithms implemented directly on the board, facilitating efficient local data processing and enabling devices to make decisions locally, rather than to rely on data streaming and latency-prone cloud computing.
<p class="heading">
<br>
<em>
<strong>
"Exploiting the potential of RRAM technology for extremely power-efficient neuromorphic systems, specifically Hardware Accelerators and Physical Neural Networks for ML applications"
</strong>
</em>
<br>
" "
</p>
听
Outputs听
- Publications
- A. Mehonic, M. Buckwell, L. Montesi, M. Munde, D. Gao, S. Hudziak, R.J. Chater, S. Fearn, D. McPhail, M. Bosman, A.L. Shluger. 鈥淣anoscale Transformations in Metastable, Amorphous, Silicon-Rich Silica." Advanced Materials 28(34), 7486-7493, (2016) doi: 10.1002/adma.201601208
- A. Mehonic and A. J Kenyon 鈥淓mulating the electrical activity of the neuron using a silicon oxide RRAM cell.鈥 Frontiers in Neuroscience 10:57, (2016) doi: 10.3389/fnins.2016.00057
- A. Mehonic, M.S. Munde, W.H. Ng, M. Buckwell, L. Montesi, M. Bosman, A.L. Shluger, A.J.Kenyon. 鈥淚ntrinsic resistance switching in amorphous silicon oxide for high performance SiOx ReRAM devices.鈥 Microelectronic Engineering.178:98-103, (2017). doi: 10.1016/j.mee.2017.04.033
- M.S. Munde, A. Mehonic , W.H Ng , M. Buckwell , L. Montesi , M. Bosman, A.L. Shluger , A. J.Kenyon 鈥淚ntrinsic Resistance Switching in Amorphous Silicon Suboxides: The Role of Columnar Microstructure.鈥, Scientific reports 7.1 9274, (2017). doi: 10.1038/s41598-017-09565-8
- A. Mehonic, T. Gerard, A. J. Kenyon. "Light-activated resistance switching in SiOx RRAM devices." Applied Physics Letters 111.23, 233502, (2017). doi: 10.1063/1.5009069
- K. Zarudnyi, A. Mehonic, L. Montesi, M. Buckwell, S. Hudziak, and A. J. Kenyon "Spike-Timing Dependent Plasticity in Unipolar Silicon Oxide RRAM Devices." Frontiers in Neuroscience 12, 57 (2018). doi: 10.3389/fnins.2018.00057