no code implementations • 25 Feb 2024 • Bo Liu, Grace Li Zhang, Xunzhao Yin, Ulf Schlichtmann, Bing Li
In this new design, the multipliers are replaced by simple logic gates to project the results onto a wide bit representation.
no code implementations • 3 Dec 2023 • Ruidi Qiu, Amro Eldebiky, Grace Li Zhang, Xunzhao Yin, Cheng Zhuo, Ulf Schlichtmann, Bing Li
In conventional ONNs, light amplitudes are modulated at the input and detected at the output.
no code implementations • 19 Sep 2023 • Kangwei Xu, Grace Li Zhang, Ulf Schlichtmann, Bing Li
However, under a given area constraint, the number of MAC units in such platforms is limited, so MAC units have to be reused to perform MAC operations in a neural network.
no code implementations • 15 Jun 2023 • Philipp van Kempen, Rafael Stahl, Daniel Mueller-Gritschneder, Ulf Schlichtmann
Thus, automating the end-to-end benchmarking flow is of high relevance nowadays.
no code implementations • 10 Jun 2023 • Chuangtao Chen, Grace Li Zhang, Xunzhao Yin, Cheng Zhuo, Ulf Schlichtmann, Bing Li
Deep neural networks (DNNs) have been widely deployed across diverse domains such as computer vision and natural language processing.
1 code implementation • 31 Mar 2023 • Rafael Stahl, Daniel Mueller-Gritschneder, Ulf Schlichtmann
It improves TinyML memory optimization significantly by reducing memory of models where this was not possible before and additionally providing alternative design points for models that show high run time overhead with existing methods.
no code implementations • 24 Mar 2023 • Richard Petri, Grace Li Zhang, Yiran Chen, Ulf Schlichtmann, Bing Li
To address this challenge, we propose PowerPruning, a novel method to reduce power consumption in digital neural network accelerators by selecting weights that lead to less power consumption in MAC operations.
no code implementations • 29 Dec 2022 • Christopher Wolters, Brady Taylor, Edward Hanson, Xiaoxuan Yang, Ulf Schlichtmann, Yiran Chen
Using the benchmarking framework DNN+NeuroSim, we investigate the impact of hardware nonidealities and quantization on algorithm performance, as well as how network topologies and algorithm-level design choices can scale latency, energy and area consumption of a chip.
no code implementations • 27 Nov 2022 • Wenhao Sun, Grace Li Zhang, Huaxi Gu, Bing Li, Ulf Schlichtmann
In the proposed method, the importance score of each filter or neuron with respect to the number of classes in the dataset is first evaluated.
no code implementations • 27 Nov 2022 • Wenhao Sun, Grace Li Zhang, Xunzhao Yin, Cheng Zhuo, Huaxi Gu, Bing Li, Ulf Schlichtmann
In such platforms, neural networks need to provide acceptable results quickly and the accuracy of the results should be able to be enhanced dynamically according to the computational resources available in the computing system.
no code implementations • 27 Nov 2022 • Amro Eldebiky, Grace Li Zhang, Georg Boecherer, Bing Li, Ulf Schlichtmann
These acceleration platforms rely on analog properties of the devices and thus suffer from process variations and noise.
no code implementations • 20 Sep 2021 • Felix Last, Ceren Yeni, Ulf Schlichtmann
As the relative power, performance, and area (PPA) impact of embedded memories continues to grow, proper parameterization of each of the thousands of memories on a chip is essential.
no code implementations • 5 Mar 2020 • Felix Last, Max Haeberlein, Ulf Schlichtmann
A key task in the design flow of a chip is to find optimal memory compiler parametrizations which on the one hand fulfill system requirements while on the other hand optimize PPA.