Search Results for author: Holger Fröning

Found 15 papers, 4 papers with code

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

no code implementations11 Jan 2024 Yannick Emonds, Kai Xi, Holger Fröning

Resistive memory is a promising alternative to SRAM, but is also an inherently unstable device that requires substantial effort to ensure correct read and write operations.

Image Classification

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

no code implementations28 Nov 2023 Daniel Barley, Holger Fröning

We report the effectiveness of activation pruning by evaluating training speed, accuracy, and memory usage of large-scale neural architectures on the example of ResMLP on image classification tasks.

Image Classification

On the Non-Associativity of Analog Computations

no code implementations25 Sep 2023 Lisa Kuhn, Bernhard Klein, Holger Fröning

With this model we assess the importance of ordering by comparing the test accuracy of a neural network for keyword spotting, which is trained based either on an ordered model, on a non-ordered variant, and on real hardware.

Keyword Spotting

Reducing Memory Requirements for the IPU using Butterfly Factorizations

no code implementations16 Sep 2023 S. -Kazem Shekofteh, Christian Alles, Holger Fröning

High Performance Computing (HPC) benefits from different improvements during last decades, specially in terms of hardware platforms to provide more processing power while maintaining the power consumption at a reasonable level.

Walking Noise: Understanding Implications of Noisy Computations on Classification Tasks

no code implementations20 Dec 2022 Hendrik Borras, Bernhard Klein, Holger Fröning

We then investigate the implications of additive and multiplicative noise for different classification tasks and model architectures, with and without batch normalization.

Towards Hardware-Specific Automatic Compression of Neural Networks

no code implementations15 Dec 2022 Torben Krieger, Bernhard Klein, Holger Fröning

Moreover, we can demonstrate that a joint search and compression using pruning and quantization is superior to an individual search for policies using a single compression method.

Quantization reinforcement-learning +1

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

no code implementations31 May 2022 Dennis Rieber, Moritz Reiber, Oliver Bringmann, Holger Fröning

From these results, a validity-driven initialization method for AutoTVM is developed, only requiring 41. 6% of the necessary hardware measurements to find the best solution, while improving search robustness.

Joint Program and Layout Transformations to enable Convolutional Operators on Specialized Hardware based on Constraint Programming

no code implementations10 Apr 2021 Dennis Rieber, Axel Acosta, Holger Fröning

First solutions to this problem have been proposed, such as TVM, UNIT or ISAMIR, which work on a loop-level representation of operators and specify data layout and possible program transformations before the embedding into the operator is performed.

Understanding Cache Boundness of ML Operators on ARM Processors

1 code implementation1 Feb 2021 Bernhard Klein, Christoph Gratl, Manfred Mücke, Holger Fröning

Machine Learning compilers like TVM allow a fast and flexible deployment on embedded CPUs.

Quantization

Resource-Efficient Speech Mask Estimation for Multi-Channel Speech Enhancement

no code implementations22 Jul 2020 Lukas Pfeifenberger, Matthias Zöhrer, Günther Schindler, Wolfgang Roth, Holger Fröning, Franz Pernkopf

While machine learning techniques are traditionally resource intensive, we are currently witnessing an increased interest in hardware and energy efficient approaches.

BIG-bench Machine Learning Speech Enhancement

On the Difficulty of Designing Processor Arrays for Deep Neural Networks

1 code implementation24 Jun 2020 Kevin Stehle, Günther Schindler, Holger Fröning

We present an analysis of popular DNN models to illustrate how it can estimate required cycles, data movement costs, as well as systolic array utilization, and show how the progress in network architecture design impacts the efficiency of inference on accelerators based on systolic arrays.

N-Ary Quantization for CNN Model Compression and Inference Acceleration

no code implementations ICLR 2019 Günther Schindler, Wolfgang Roth, Franz Pernkopf, Holger Fröning

In this work we propose a method for weight and activation quantization that is scalable in terms of quantization levels (n-ary representations) and easy to compute while maintaining the performance close to full-precision CNNs.

Clustering Model Compression +1

Cannot find the paper you are looking for? You can Submit a new open access paper.