Login / Signup
Echo: Compiler-based GPU Memory Footprint Reduction for LSTM RNN Training.
Bojian Zheng
Nandita Vijaykumar
Gennady Pekhimenko
Published in:
ISCA (2020)
Keyphrases
</>
memory footprint
recurrent neural networks
long short term memory
memory usage
programming language
training phase
feed forward
training algorithm
online learning
supervised learning
data sets
nearest neighbor
neural network
high frequency
software systems
training data
gpu implementation
spectrum analysis