Alert button
Picture for Holger Fröning

Holger Fröning

Alert button

Implications of Noise in Resistive Memory on Deep Neural Networks for Image Classification

Jan 11, 2024
Yannick Emonds, Kai Xi, Holger Fröning

Viaarxiv icon

Compressing the Backward Pass of Large-Scale Neural Architectures by Structured Activation Pruning

Nov 29, 2023
Daniel Barley, Holger Fröning

Viaarxiv icon

On the Non-Associativity of Analog Computations

Sep 25, 2023
Lisa Kuhn, Bernhard Klein, Holger Fröning

Viaarxiv icon

Reducing Memory Requirements for the IPU using Butterfly Factorizations

Sep 16, 2023
S. -Kazem Shekofteh, Christian Alles, Holger Fröning

Viaarxiv icon

Walking Noise: Understanding Implications of Noisy Computations on Classification Tasks

Dec 20, 2022
Hendrik Borras, Bernhard Klein, Holger Fröning

Figure 1 for Walking Noise: Understanding Implications of Noisy Computations on Classification Tasks
Figure 2 for Walking Noise: Understanding Implications of Noisy Computations on Classification Tasks
Figure 3 for Walking Noise: Understanding Implications of Noisy Computations on Classification Tasks
Figure 4 for Walking Noise: Understanding Implications of Noisy Computations on Classification Tasks
Viaarxiv icon

Towards Hardware-Specific Automatic Compression of Neural Networks

Dec 15, 2022
Torben Krieger, Bernhard Klein, Holger Fröning

Figure 1 for Towards Hardware-Specific Automatic Compression of Neural Networks
Figure 2 for Towards Hardware-Specific Automatic Compression of Neural Networks
Figure 3 for Towards Hardware-Specific Automatic Compression of Neural Networks
Figure 4 for Towards Hardware-Specific Automatic Compression of Neural Networks
Viaarxiv icon

HW-Aware Initialization of DNN Auto-Tuning to Improve Exploration Time and Robustness

May 31, 2022
Dennis Rieber, Moritz Reiber, Oliver Bringmann, Holger Fröning

Figure 1 for HW-Aware Initialization of DNN Auto-Tuning to Improve Exploration Time and Robustness
Figure 2 for HW-Aware Initialization of DNN Auto-Tuning to Improve Exploration Time and Robustness
Figure 3 for HW-Aware Initialization of DNN Auto-Tuning to Improve Exploration Time and Robustness
Figure 4 for HW-Aware Initialization of DNN Auto-Tuning to Improve Exploration Time and Robustness
Viaarxiv icon

The Programming of Deep Learning Accelerators as a Constraint Satisfaction Problem

Apr 13, 2021
Dennis Rieber, Axel Acosta, Holger Fröning

Figure 1 for The Programming of Deep Learning Accelerators as a Constraint Satisfaction Problem
Figure 2 for The Programming of Deep Learning Accelerators as a Constraint Satisfaction Problem
Figure 3 for The Programming of Deep Learning Accelerators as a Constraint Satisfaction Problem
Figure 4 for The Programming of Deep Learning Accelerators as a Constraint Satisfaction Problem
Viaarxiv icon

Understanding Cache Boundness of ML Operators on ARM Processors

Feb 01, 2021
Bernhard Klein, Christoph Gratl, Manfred Mücke, Holger Fröning

Figure 1 for Understanding Cache Boundness of ML Operators on ARM Processors
Figure 2 for Understanding Cache Boundness of ML Operators on ARM Processors
Figure 3 for Understanding Cache Boundness of ML Operators on ARM Processors
Figure 4 for Understanding Cache Boundness of ML Operators on ARM Processors
Viaarxiv icon