high performance computing on graphics processing units: hgpu.org

hgpu.org » Machine learning

Machine Learning Based Auto-tuning for Enhanced OpenCL Performance Portability

Thomas L. Falch, Anne C. Elster

View

Tags: ATI, ATI Radeon HD 7970, Benchmarking, Computer science, Deep learning, Heterogeneous systems, Machine learning, Neural networks, nVidia, OpenCL, Tesla K40

June 5, 2015 by hgpu

Blocks and Fuel: Frameworks for deep learning

Bart van Merrienboer, Dzmitry Bahdanau, Vincent Dumoulin, Dmitriy Serdyuk, David Warde-Farley, Jan Chorowski, Yoshua Bengio

View

Download (PDF)

Source codes

Tags: Computer science, CUDA, Deep learning, Linear Algebra, Machine learning, MPI, Neural networks, nVidia, Package, Python, Symbolic Computation, Visualization

June 5, 2015 by hgpu

DjiNN and Tonic: DNN as a Service and Its Implications for Future Warehouse Scale Computers

Johann Hauswald, Yiping Kang, Michael A. Laurenzano, Quan Chen, Cheng Li, Trevor Mudge, Ronald G. Dreslinski, Jason Mars, Lingjia Tang

View

Download (PDF)

Tags: CUDA, Deep learning, Image processing, Machine learning, Neural networks, nVidia, Speech recognition, Tesla K40

May 25, 2015 by hgpu

ACCTuner: OpenACC Auto-Tuner For Accelerated Scientific Applications

Fatemah Ramzy AlZayer

View

Download (PDF)

Tags: Algorithms, Computer science, Machine learning, Nearest neighbour, nVidia, OpenACC, Tesla K20, Tesla K40, Thesis

May 25, 2015 by hgpu

Using Butterfly-Patterned Partial Sums to Optimize GPU Memory Accesses for Drawing from Discrete Distributions

Guy L. Steele Jr. (Oracle Labs), Jean-Baptiste Tristan

View

Download (PDF)

Tags: Computer science, CUDA, Latent Dirichlet allocation, Machine learning, nVidia, nVidia GeForce Titan

May 16, 2015 by hgpu

Speeding up Automatic Hyperparameter Optimization of Deep Neural Networks by Extrapolation of Learning Curves

Tobias Domhan, Jost Tobias Springenberg, Frank Hutter

View

Download (PDF)

Tags: Benchmarking, Computer science, CUDA, Deep learning, Machine learning, Neural networks, nVidia, nVidia GeForce GTX Titan

May 15, 2015 by hgpu

PerforatedCNNs: Acceleration through Elimination of Redundant Convolutions

Michael Figurnov, Dmitry Vetrov, Pushmeet Kohli

View

Download (PDF)

Tags: CNN, Computer science, CUDA, Deep learning, Machine learning, Neural networks, nVidia, nVidia GeForce GTX 980

May 5, 2015 by hgpu

Caffe con Troll: Shallow Ideas to Speed Up Deep Learning

Firas Abuzaid, Stefan Hadjis, Ce Zhang, Christopher Re

View

Download (PDF)

Tags: Computer science, CUDA, Deep learning, Machine learning, Neural networks, nVidia, nVidia GRID K520, Tesla K40

April 20, 2015 by hgpu

Image Classification with Pyramid Representation and Rotated Data Augmentation on Torch 7

Keven (Kedao) Wang

View

Download (PDF)

Tags: Computer science, CUDA, Deep learning, Machine learning, Neural networks, nVidia, nVidia GRID K520

April 14, 2015 by hgpu

Multi-Lingual Speech Recognition with Low-Rank Multi-Task Deep Neural Networks

Aanchan Mohan, Richard Rose

View

Download (PDF)

Tags: Computer science, CUDA, Deep learning, Machine learning, Neural networks, nVidia, Speech recognition, Tesla K20

April 7, 2015 by hgpu

Face Retriever: Pre-filtering the Gallery via Deep Neural Net

Dayong Wang, Anil K. Jain

View

Download (PDF)

Tags: Algorithms, Computer science, Computer vision, CUDA, Deep learning, Filtering, Machine learning, Neural networks, nVidia, Tesla K20

March 30, 2015 by hgpu

Single stream parallelization of generalized LSTM-like RNNs on a GPU

Kyuyeon Hwang, Wonyong Sung

View

Download (PDF)

Tags: Algorithms, Computer science, CUDA, Machine learning, Neural and Evolutionary Computing, Neural networks, nVidia, Tesla K40

March 22, 2015 by hgpu

MSCCL++: A GPU-driven communication stack for scalable AI applications

MSCCL++: Rethinking GPU Communication Abstractions for Cutting-edge AI Applications

Benchmark compute shader of Unity against InteropUnityCUDA

InteropUnityCUDA: A Tool for Interoperability Between Unity and CUDA

Data-efficient LLM Fine-tuning for Code Generation

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

DeepCompile: A Compiler-Driven Approach to Optimizing Distributed Deep Learning Training

Large Language Model Powered C-to-CUDA Code Translation: A Novel Auto-Parallelization Framework

GigaAPI: a user-space API that simplifies multi-GPU programming, bridging the gap between the capabilities of parallel GPU systems and the ability of developers to harness their full potential

GigaAPI for GPU Parallelization

Coccinelle: a C code transformation engine using SmPL for matches, refactorings, and bug fixing

Advances in Semantic Patching for HPC-oriented Refactorings with Coccinelle

* * *

high performance computing on graphics processing units: hgpu.org

Machine Learning Based Auto-tuning for Enhanced OpenCL Performance Portability

Blocks and Fuel: Frameworks for deep learning

DjiNN and Tonic: DNN as a Service and Its Implications for Future Warehouse Scale Computers

ACCTuner: OpenACC Auto-Tuner For Accelerated Scientific Applications

Using Butterfly-Patterned Partial Sums to Optimize GPU Memory Accesses for Drawing from Discrete Distributions

Speeding up Automatic Hyperparameter Optimization of Deep Neural Networks by Extrapolation of Learning Curves

PerforatedCNNs: Acceleration through Elimination of Redundant Convolutions

Caffe con Troll: Shallow Ideas to Speed Up Deep Learning

Image Classification with Pyramid Representation and Rotated Data Augmentation on Torch 7

Multi-Lingual Speech Recognition with Low-Rank Multi-Task Deep Neural Networks

Face Retriever: Pre-filtering the Gallery via Deep Neural Net

Single stream parallelization of generalized LSTM-like RNNs on a GPU

Recent source codes

MSCCL++: A GPU-driven communication stack for scalable AI applications

Benchmark compute shader of Unity against InteropUnityCUDA

Data-efficient LLM Fine-tuning for Code Generation

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Large Language Model Powered C-to-CUDA Code Translation: A Novel Auto-Parallelization Framework

GigaAPI: a user-space API that simplifies multi-GPU programming, bridging the gap between the capabilities of parallel GPU systems and the ability of developers to harness their full potential

Coccinelle: a C code transformation engine using SmPL for matches, refactorings, and bug fixing

DuoReduce: MLIR's benchmark

Shamrock: Multi-GPU hydrodynamics for astrophysics

LLMPerf: GPU Performance Modeling meets Large Language Models

Most viewed papers (last 30 days)