research
∙
10/19/2022
Tempo: Accelerating Transformer-Based Model Training through Memory Footprint Reduction
Training deep learning models can be computationally expensive. Prior wo...
share
research
∙
10/18/2022
Hidet: Task-Mapping Programming Paradigm for Deep Learning Tensor Programs
As deep learning models nowadays are widely adopted by both cloud servic...
research
∙
07/02/2020
Automatic Horizontal Fusion for GPU Kernels
We present automatic horizontal fusion, a novel optimization technique t...
research
∙
05/22/2018
EcoRNN: Fused LSTM RNN Implementation with Data Layout Optimization
Long-Short-Term-Memory Recurrent Neural Network (LSTM RNN) is a state-of...
research
∙
03/16/2018