Deep Learning

Technical Review on PyTorch 2.0 and Triton
High-level overview of PyTorch 2.0 and Triton integration
Aug 7, 2023 10:03 PM — 10:03 PM Virtual
Project Slides
Hardware-Aware Compression with Random Operation Access Specific Tile (ROAST) Hashing
Advancements in deep learning are often associated with increasing model sizes. Training and deploying large models require …
Cite Project URL
Towards Agile Development of Efficient Deep Learning Operators (Hardware Insights)
Presented a talk about Triton and requested feedback from Intel engineers
Jun 29, 2023 10:56 PM — 10:56 PM Virtual
Project Slides
Towards Agile Development of Efficient Deep Learning Operators (Call for Contributions)
Presented a talk about Triton and called for contributions to improving the language
Jun 19, 2023 10:56 PM — 10:56 PM Lake Tahoe, California
Project Slides
Paw-Net: Stacking Ensemble Deep Learning for Segmenting Scanning Electron Microscopy Images of Fine-grained Shale Samples
Segmentation of scanning electron microscopy (SEM) images is critical yet time-consuming for geological analyses, as it needs to …
Cite DOI URL
A Performance Analysis Framework for Exploiting GPU Microarchitectural Capability
Presented our work on static performance analysis for GPUs at ICS17
Jul 20, 2017 9:36 PM — 9:36 PM Chicago, IL, USA
Slides
Deep Learning on Modern Architectures
Discussed how state-of-the-art deep learning libraries optimize computations by utilizing architectural features.
Apr 1, 2017 10:00 PM — 10:00 PM Institute of Computing Technology, Chinese Academy of Sciences
Slides
Convolution Methods
Introduced various kinds of convolution methods and analyzed their complexities, memory consumptions, and data access patterns.
Jun 1, 2016 9:58 PM — 9:58 PM Institute of Computing Technology, Chinese Academy of Sciences
Slides