Long Live TIME: Improving Lifetime and Security for NVM-Based Training-in-Memory Systems
Nonvolatile memory (NVM)-based training-in-memory (TIME) systems have emerged that can process the neural network (NN) training in an energy-efficient manner. However, the endurance of NVM cells is disappointing, rendering concerns about the lifetime of TIME systems, because the weights of NN models...
Príomhchruthaitheoirí: | Lin, Yujun, Han, Song |
---|---|
Rannpháirtithe: | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
Formáid: | Alt |
Foilsithe / Cruthaithe: |
IEEE
2021
|
Rochtain ar líne: | https://hdl.handle.net/1721.1/129440 |
Míreanna comhchosúla
Míreanna comhchosúla
-
Long Live TIME: Improving Lifetime for Training-In-Memory Engines by Structured Gradient Sparsification
de réir: Han, Song
Foilsithe / Cruthaithe: (2021) -
Si nanowire based NVM : SONOS fabrication & characterization
de réir: Chen, Mincong.
Foilsithe / Cruthaithe: (2009) -
Next generation programming environments and architectures for NVM-based computing and storage systems
de réir: Poh, Shie Liang
Foilsithe / Cruthaithe: (2015) -
Design of non-destructive single-sawtooth pulse based readout for STT-RAM by NVM-SPICE
de réir: Wang, Yuhao, et al.
Foilsithe / Cruthaithe: (2013) -
A novel FPGA implementation of mirror-paradigm RS-based QC-LDPC decoder for NVM channels
de réir: Lim, Melvin Heng Li, et al.
Foilsithe / Cruthaithe: (2013)