
Characterizing and Demystifying the Implicit Convolution Algorithm on Commercial MatrixMultiplication Accelerators
Many of today's deep neural network accelerators, e.g., Google's TPU and...
read it

Dubhe: Towards Data Unbiasedness with Homomorphic Encryption in Federated Learning Client Selection
Federated learning (FL) is a distributed machine learning paradigm that ...
read it

ZIPPER: Exploiting Tile and Operatorlevel Parallelism for General and Scalable Graph Neural Network Acceleration
Graph neural networks (GNNs) start to gain momentum after showing signif...
read it

Dualside Sparse Tensor Core
Leveraging sparsity in deep neural network (DNN) models is promising for...
read it

DLFusion: An AutoTuning Compiler for Layer Fusion on Deep Neural Network Accelerator
Many hardware vendors have introduced specialized deep neural networks (...
read it

How Far Does BERT Look At:Distancebased Clustering and Analysis of BERT's Attention
Recent research on the multihead attention mechanism, especially that i...
read it

Architectural Implications of Graph Neural Networks
Graph neural networks (GNN) represent an emerging line of deep learning ...
read it

Accelerating Sparse DNN Models without HardwareSupport via TileWise Sparsity
Network pruning can reduce the high computation cost of deep neural netw...
read it

Ptolemy: Architecture Support for Robust Deep Learning
Deep learning is vulnerable to adversarial attacks, where carefullycraf...
read it

Exceeding Conservative Limits: A Consolidated Analysis on Modern Hardware Margins
Modern largescale computing systems (data centers, supercomputers, clou...
read it

Towards QoSAware and ResourceEfficient GPU Microservices Based on Spatial Multitasking GPUs In Datacenters
While prior researches focus on CPUbased microservices, they are not ap...
read it

Balancing Efficiency and Flexibility for DNN Acceleration via Temporal GPUSystolic Array Integration
The research interest in specialized hardware accelerators for deep neur...
read it

Adversarial Defense Through Network Profiling Based Path Extraction
Recently, researchers have started decomposing deep neural network model...
read it
Jingwen Leng
is this you? claim profile