11 citations · 24 across the 41 of their papers we have counts for
7 papers · 1 filter
AMPED: Accelerating MTTKRP for Billion-Scale Sparse Tensor Decomposition on Multiple GPUs
Sasindu Wijeratne, Rajgopal Kannan, Viktor Prasanna
Matricized Tensor Times Khatri-Rao Product (MTTKRP) is the computational bottleneck in sparse tensor decomposition. As real-world sparse tensors grow to billions of nonzeros, they…
Accelerating Sparse MTTKRP for Small Tensor Decomposition on GPU
Sasindu Wijeratne, Rajgopal Kannan, Viktor Prasanna
Sparse Matricized Tensor Times Khatri-Rao Product (spMTTKRP) is the bottleneck kernel of sparse tensor decomposition. In tensor decomposition, spMTTKRP is performed iteratively alo…
Sparse MTTKRP Acceleration for Tensor Decomposition on GPU
Sasindu Wijeratne, Rajgopal Kannan, Viktor Prasanna
Sparse Matricized Tensor Times Khatri-Rao Product (spMTTKRP) is the bottleneck kernel of sparse tensor decomposition. In this work, we propose a GPU-based algorithm design to addre…
Accelerating ViT Inference on FPGA through Static and Dynamic Pruning
Dhruv Parikh, Shouyi Li, Bingyi Zhang +3
Vision Transformers (ViTs) have achieved state-of-the-art accuracy on various computer vision tasks. However, their high computational complexity prevents them from being applied t…
GCV-Turbo: End-to-end Acceleration of GNN-based Computer Vision Tasks on FPGA
Bingyi Zhang, Rajgopal Kannan, Carl Busart +1
Graph neural networks (GNNs) have recently empowered various novel computer vision (CV) tasks. In GNN-based CV tasks, a combination of CNN layers and GNN layers or only GNN layers…
Performance of Graph Neural Networks for Point Cloud Applications
Dhruv Parikh, Bingyi Zhang, Rajgopal Kannan +2
Graph Neural Networks (GNNs) have gained significant momentum recently due to their capability to learn on unstructured graph data. Dynamic GNNs (DGNNs) are the current state-of-th…
GPOP: A cache- and work-efficient framework for Graph Processing Over Partitions
Kartik Lakhotia, Sourav Pati, Rajgopal Kannan +1
Past decade has seen the development of many shared-memory graph processing frameworks, intended to reduce the effort of developing high performance parallel applications. However…