Login / Signup
TensorDIMM: A Practical Near-Memory Processing Architecture for Embeddings and Tensor Operations in Deep Learning.
Youngeun Kwon
Yunjae Lee
Minsoo Rhu
Published in:
MICRO (2019)
Keyphrases
</>
deep learning
processing elements
processing units
memory management
unsupervised feature learning
machine learning
unsupervised learning
higher order
dimensionality reduction
mental models
weakly supervised
feature extraction
deep architectures
denoising
high dimensional
natural language
computer vision