Figure 4 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Relative accuracy for Design A (top) and Design B (bottom) as function of copies on LeNet5 trained for MNIST classification. The pale area contains the 56.5%-confidence intervals.
Source publication
All analog signal processing is fundamentally subject to noise, and this is also the case in modern implementations of Optical Neural Networks (ONNs). Therefore, to mitigate noise in ONNs, we propose two designs that are constructed from a given, possibly trained, Neural Network (NN) that one wishes to implement. Both designs have the capability th...
Contexts in source publication
Context 1
... 3 displays the MSE seen for LeNet, depending on the amount of copies for each design. In the trade-off between additional resources needed for the additional copies against the diminishing benefits of adding further copies, we see that, for both measures MSE (Figure 3) and relative accuracy (Figure 4), already 2 to 5 copies per layer yield good results. The relative accuracy in Figure 4 is scaled such that 0 corresponds to the accuracy of the original NN with noise profile (i.e., the ONN without modifications, we call this the original ONN) and 1 to the accuracy of the original NN without noise. ...
Context 2
... the trade-off between additional resources needed for the additional copies against the diminishing benefits of adding further copies, we see that, for both measures MSE (Figure 3) and relative accuracy (Figure 4), already 2 to 5 copies per layer yield good results. The relative accuracy in Figure 4 is scaled such that 0 corresponds to the accuracy of the original NN with noise profile (i.e., the ONN without modifications, we call this the original ONN) and 1 to the accuracy of the original NN without noise. The designs do not alter the fundamental operation of the original NN, therefore there should be no performance gain and the original NN's accuracy should be considered the highest achievable, thus constituting the upper bound in relative accuracy of 1. Likewise the lowest accuracy should be given by the original ONN, as there is no noise reduction involved. ...
Similar publications
![Figure 1: Aoluguya Ewenki traditional women's robe [1].](publication/382425811/figure/fig1/AS:11431281261373627@1721524584173/Aoluguya-Ewenki-traditional-womens-robe-1_Q320.jpg)
![Figure 3: Aoluguya Ewenki ethnic "birch bark culture" [4].](publication/382425811/figure/fig2/AS:11431281261346744@1721524584324/Aoluguya-Ewenki-ethnic-birch-bark-culture-4_Q320.jpg)
This paper explores the innovative design and cultural heritage of traditional costumes of the Aoluguya Ewenki ethnic group in the digital age. Through digital techniques, traditional costumes have been infused with modern elements, enhancing the intricacy and diversity of their designs, thus introducing new means of design for traditional attire....
Mechanical metamaterials enable the creation of structural materials with unprecedented mechanical properties. However, thus far, research on mechanical metamaterials has focused on passive mechanical metamaterials and the tunability of their mechanical properties. Deep integration of multifunctionality, sensing, electrical actuation, information p...
The integration of Large Language Models (LLMs) into autonomous driving systems demonstrates strong common sense and reasoning abilities, effectively addressing the pitfalls of purely data-driven methods. Current LLM-based agents require lengthy inference times and face challenges in interacting with real-time autonomous driving environments. A key...
All analog signal processing is fundamentally subject to noise, and this is also the case in next generation implementations of optical neural networks (ONNs). Therefore, we propose the first hardware-based approach to mitigate noise in ONNs. A tree-like and an accordion-like design are constructed from a given NN that one wishes to implement. Both...
Offline model-based optimization (MBO) aims to maximize a black-box objective function using only an offline dataset of designs and scores. A prevalent approach involves training a conditional generative model on existing designs and their associated scores, followed by the generation of new designs conditioned on higher target scores. However, the...
