Concepedia

Accelerating machine learning with Non-Volatile Memory: Exploring device and circuit tradeoffs

Alessandro Fumarola, Pritish Narayanan, Lucas L. Sanches, Severin Sidler, Junwoo Jang, Kibong Moon, R. M. Shelby, Hyunsang Hwang, Geoffrey W. Burr

2016 · 34 citations · 7 references

DOIFull text

Open access

Concepts

Abstract

Large arrays of the same nonvolatile memories (NVM) being developed for Storage-Class Memory (SCM) - such as Phase Change Memory (PCM) and Resistance RAM (ReRAM) - can also be used in non-Von Neumann neuromorphic computational schemes, with device conductance serving as synaptic “weight.” This allows the all-important multiply-accumulate operation within these algorithms to be performed efficiently at the weight data.

References

7

Gradient-based learning applied to document recognition

Yann LeCun, Léon Bottou, Yoshua Bengio et al. · Proceedings of the IEEE · 1998 · 58.2K citations · Full text

, , +14

Experimental Demonstration and Tolerancing of a Large-Scale Neural Network (165 000 Synapses) Using Phase-Change Memory as the Synaptic Weight Element

Geoffrey W. Burr, R. M. Shelby, Severin Sidler et al. · IEEE Transactions on Electron Devices · 2015 · 923 citations · Full text

, , +16

Low-field amorphous state resistance and threshold voltage drift in chalcogenide materials

A. Pirovano, Andrea L. Lacaita, F. Pellizzer et al. · IEEE Transactions on Electron Devices · 2004 · 369 citations

, , +17

Optimization of Conductance Change in Pr<sub>1–<i>x</i></sub>Ca<sub><i>x</i></sub>MnO<sub>3</sub>-Based Synaptic Devices for Neuromorphic Systems

Sangsu Park, Geoffrey W. Burr, Hyunsang Hwang et al. · IEEE Electron Device Letters · 2015 · 306 citations

, , +13

Experimental demonstration and tolerancing of a large-scale neural network (165,000 synapses), using phase-change memory as the synaptic weight element

Geoffrey W. Burr, R. M. Shelby, Carmelo di Nolfo et al. · 2014 · 204 citations

, , +14