Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox
Picture for Torsten Hoefler

Torsten Hoefler
Alert button

ASDL: A Unified Interface for Gradient Preconditioning in PyTorch


May 08, 2023
Kazuki Osawa, Satoki Ishikawa, Rio Yokota, Shigang Li, Torsten Hoefler

Add code
Alert button


   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

STen: Productive and Efficient Sparsity in PyTorch


Apr 15, 2023
Andrei Ivanov, Nikoli Dryden, Tal Ben-Nun, Saleh Ashkboos, Torsten Hoefler

Add code
Alert button

Figure 1 for STen: Productive and Efficient Sparsity in PyTorch
Figure 2 for STen: Productive and Efficient Sparsity in PyTorch
Figure 3 for STen: Productive and Efficient Sparsity in PyTorch
Figure 4 for STen: Productive and Efficient Sparsity in PyTorch

   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Performance Embeddings: A Similarity-based Approach to Automatic Performance Optimization


Mar 14, 2023
Lukas Trümper, Tal Ben-Nun, Philipp Schaad, Alexandru Calotoiu, Torsten Hoefler

Add code
Alert button

Figure 1 for Performance Embeddings: A Similarity-based Approach to Automatic Performance Optimization
Figure 2 for Performance Embeddings: A Similarity-based Approach to Automatic Performance Optimization
Figure 3 for Performance Embeddings: A Similarity-based Approach to Automatic Performance Optimization
Figure 4 for Performance Embeddings: A Similarity-based Approach to Automatic Performance Optimization

   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Myths and Legends in High-Performance Computing


Jan 06, 2023
Satoshi Matsuoka, Jens Domke, Mohamed Wahib, Aleksandr Drozd, Torsten Hoefler

Add code
Alert button

Figure 1 for Myths and Legends in High-Performance Computing
Figure 2 for Myths and Legends in High-Performance Computing
Figure 3 for Myths and Legends in High-Performance Computing

   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

A Theory of I/O-Efficient Sparse Neural Network Inference


Jan 03, 2023
Niels Gleinig, Tal Ben-Nun, Torsten Hoefler

Add code
Alert button

Figure 1 for A Theory of I/O-Efficient Sparse Neural Network Inference
Figure 2 for A Theory of I/O-Efficient Sparse Neural Network Inference
Figure 3 for A Theory of I/O-Efficient Sparse Neural Network Inference
Figure 4 for A Theory of I/O-Efficient Sparse Neural Network Inference

   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

PipeFisher: Efficient Training of Large Language Models Using Pipelining and Fisher Information Matrices


Nov 25, 2022
Kazuki Osawa, Shigang Li, Torsten Hoefler

Add code
Alert button

Figure 1 for PipeFisher: Efficient Training of Large Language Models Using Pipelining and Fisher Information Matrices
Figure 2 for PipeFisher: Efficient Training of Large Language Models Using Pipelining and Fisher Information Matrices
Figure 3 for PipeFisher: Efficient Training of Large Language Models Using Pipelining and Fisher Information Matrices
Figure 4 for PipeFisher: Efficient Training of Large Language Models Using Pipelining and Fisher Information Matrices

   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Spatial Mixture-of-Experts


Nov 24, 2022
Nikoli Dryden, Torsten Hoefler

Add code
Alert button

Figure 1 for Spatial Mixture-of-Experts
Figure 2 for Spatial Mixture-of-Experts
Figure 3 for Spatial Mixture-of-Experts
Figure 4 for Spatial Mixture-of-Experts
* 20 pages, 3 figures; NeurIPS 2022 
   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers


Oct 31, 2022
Elias Frantar, Saleh Ashkboos, Torsten Hoefler, Dan Alistarh

Add code
Alert button

Figure 1 for GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers
Figure 2 for GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers
Figure 3 for GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers
Figure 4 for GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers

   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Compressing multidimensional weather and climate data into neural networks


Oct 22, 2022
Langwen Huang, Torsten Hoefler

Add code
Alert button

Figure 1 for Compressing multidimensional weather and climate data into neural networks
Figure 2 for Compressing multidimensional weather and climate data into neural networks
Figure 3 for Compressing multidimensional weather and climate data into neural networks
Figure 4 for Compressing multidimensional weather and climate data into neural networks

   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Neural Graph Databases


Sep 20, 2022
Maciej Besta, Patrick Iff, Florian Scheidl, Kazuki Osawa, Nikoli Dryden, Michal Podstawski, Tiancheng Chen, Torsten Hoefler

Add code
Alert button

Figure 1 for Neural Graph Databases
Figure 2 for Neural Graph Databases
Figure 3 for Neural Graph Databases
Figure 4 for Neural Graph Databases

   Access Paper or Ask Questions
  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email
1
2
3
4
5
>>