464,215 Hits in 2.7 sec

Physics-based Deep Learning [article]

Nils Thuerey and Philipp Holl and Maximilian Mueller and Patrick Schnell and Felix Trost and Kiwon Um
2021 arXiv   pre-print
This digital book contains a practical and comprehensive introduction of everything related to deep learning in the context of physical simulations.  ...  Beyond standard supervised learning from data, we'll look at physical loss constraints, more tightly coupled learning algorithms with differentiable simulations, as well as reinforcement learning and uncertainty  ...  191 Physics-based Deep Learning 192 Chapter 20.  ... 
arXiv:2109.05237v2 fatcat:dm2wyckg6fcxzhsxi4hmo76sny

Physics Based Deep Learning Technique for Prognostics

Khaled Akkad
2019 Proceedings of the Annual Conference of the Prognostics and Health Management Society, PHM  
Various deep learning (DL) based techniques have been developed and applied for the purposes of RUL estimation.  ...  The contribution of the research relies on creating hybrid DL based techniques as well as combining physics based approaches with DL techniques for effective RUL prediction.  ...  The logical next step is to find more suitable combinations of DL based approaches using meta-learning.  ... 
doi:10.36001/phmconf.2019.v11i1.916 fatcat:3j6boyun3naajmtuxbfyskzqyy

Deep Learning based Algorithms in Astroparticle Physics

Martin Erdmann, Jonas Glombitza
2020 Journal of Physics, Conference Series  
We summarize the latest results, discuss the algorithms and challenges and further illustrate the opportunities for the astrophysics community offered by deep learning based algorithms.  ...  In recent years, great progress has been made in the fields of machine translation, image classification and speech recognition by using deep neural networks and associated techniques (deep learning).  ...  Data In elementary particle physics, many deep learning algorithms are based on the processing of the 4-momenta of measured particles in the respective particle detector.  ... 
doi:10.18154/rwth-2021-02722 fatcat:n7k2z2gbijbell7y7lcguvuf44

Simulation free reliability analysis: A physics-informed deep learning based approach [article]

Souvik Chakraborty
2020 arXiv   pre-print
The method proposed is rooted in a recently developed deep learning approach, referred to as the physics-informed neural network.  ...  The primary idea is to learn the neural network parameters directly from the physics of the problem. With this, the need for running simulation and generating data is completely eliminated.  ...  To address this issue, the concept of physics-informed deep learning was proposed in [44] .  ... 
arXiv:2005.01302v3 fatcat:uvoc6qpionde7ao72csppyf4oy

Transfer learning based multi-fidelity physics informed deep neural network [article]

Souvik Chakraborty
2020 arXiv   pre-print
MF-PIDNN blends physics informed and data-driven deep learning techniques by using the concept of transfer learning.  ...  The approximate governing equation is first used to train a low-fidelity physics informed deep neural network.  ...  MF-PIDNN blends the concepts of physics-informed and data-driven deep learning; the primary idea is to first train a low-fidelity deep learning model based on the available approximate physics and then  ... 
arXiv:2005.10614v2 fatcat:3oovcmxgyjbivc56dt57bn2zdm

Adversarial camera stickers: A physical camera-based attack on deep learning systems [article]

Juncheng Li, Frank R. Schmidt, J. Zico Kolter
2019 arXiv   pre-print
Recent work has documented the susceptibility of deep learning systems to adversarial examples, but most such attacks directly manipulate the digital input to a classifier.  ...  In this work, we consider an alternative question: is it possible to fool deep classifiers, over all perceived objects of a certain type, by physically manipulating the camera itself?  ...  Introduction Many recent papers have established that deep learning classifiers are particularly susceptible to adversarial attacks, manipulations of the input to a classifier specifically crafted to  ... 
arXiv:1904.00759v4 fatcat:lovrtm7ejjc3tlgnhsrdtmj5ua

Fusing Physics-based and Deep Learning Models for Prognostics [article]

Manuel Arias Chao, Chetan Kulkarni, Kai Goebel, Olga Fink
2020 arXiv   pre-print
Combining the advantages of these two directions while overcoming some of their limitations, we propose a novel hybrid framework for fusing the information from physics-based performance models with deep  ...  of physics-based models and (2) limited representativeness of the training dataset for data-driven models.  ...  In particular, to benefit from the learning ability of recent advances in deep learning, we propose to combine the physics-based performance models with deep learning architectures.  ... 
arXiv:2003.00732v2 fatcat:icqekdh2nne6nmaqhjhcfdm5ya

Physics-Based Deep Learning for Fiber-Optic Communication Systems [article]

Christian Häger, Henry D. Pfister
2020 arXiv   pre-print
The resulting physics-based machine-learning model has several advantages over "black-box" function approximators.  ...  Rather than employing neural networks, the proposed algorithm, dubbed learned DBP (LDBP), uses the physics-based model with trainable filters in each step and its complexity is reduced by progressively  ...  The resulting physics-based model was shown to have a similar mathematical structure compared to standard deep neural networks.  ... 
arXiv:2010.14258v1 fatcat:vpee25gvszgbjn3ivcudubp5qu

Single Plane-Wave Imaging using Physics-Based Deep Learning [article]

Georgios Pilikos, Chris L. de Korte, Tristan van Leeuwen, Felix Lucka
2021 arXiv   pre-print
This illustrates the great potential of combining deep neural networks with physics-based image formation algorithms for SPW imaging.  ...  Here, we propose a data-to-image architecture that incorporates a wave-physics-based image formation algorithm in-between deep convolutional neural networks.  ...  Figure 1 illustrates our proposed physics-based deep learning architecture. IV.  ... 
arXiv:2109.03661v1 fatcat:n6fdwetdozalledkgz6hepadky

Physical deep learning based on optimal control of dynamical systems [article]

Genki Furuhata, Tomoaki Niiyama, Satoshi Sunada
2020 arXiv   pre-print
The learning is based on the adjoint method to optimally control dynamical systems, and the deep (virtual) network structures based on the time evolution of the systems can be used for processing input  ...  Here, we present a pattern recognition based on optimal control of continuous-time dynamical systems, which is suitable for physical hardware implementation.  ...  that there is a connection between deep neural networks and dynamical systems and indicates the possibility of using dynamical systems as physical deep-learning machines.  ... 
arXiv:2012.08761v1 fatcat:cl2fuzwoenezxjno3hahkv3nzu

Combining physics-based modeling and deep learning for ultrasound elastography [article]

Narges Mohammadi, Marvin M. Doyley, Mujdat Cetin
2021 arXiv   pre-print
We present an integrated objective function composed of a statistical physics-based forward model and a data-driven regularizer to leverage deep neural networks for learning the underlying elasticity prior  ...  In this regard, we propose a joint model-based and learning-based framework for estimating the elasticity distribution by solving a regularized optimization problem.  ...  One group of methods for integrating physics-based modeling and deep learning priors including Plug-and-Play (PnP) [8, 9, 10, 11] and regularization by denoising (RED) [12] seeks to learn a data-adaptive  ... 
arXiv:2107.13120v1 fatcat:qa6jqv45hzagva3r4mh5oqfgbq

Cyber-Physical System for Environmental Monitoring Based on Deep Learning

Íñigo Monedero, Julio Barbancho, Rafael Márquez, Juan F. Beltrán
2021 Sensors  
This paper proposes a deep learning classification sound system for execution over CPS.  ...  Cyber-physical systems (CPS) constitute a promising paradigm that could fit various applications.  ...  The paper [19] uses two deep-learning methods that apply convolutional neural networks to anuran classification.  ... 
doi:10.3390/s21113655 pmid:34073979 fatcat:drsi4a7prnantlko2scxvcr7jy

Solar Filament Recognition Based on Deep Learning

Gaofei Zhu, Ganghua Lin, Dongguang Wang, Suo Liu, Xiao Yang
2019 Solar Physics  
Firstly, a raw filament dataset is set up, consisting of tens of thousands of images required for deep learning.  ...  The paper presents a reliable method using deep learning to recognize solar filaments in H-alpha full-disk solar images automatically.  ...  Hinton & Salakhutdinov (2006) formally proposed the concept of deep learning with two viewpoints.  ... 
doi:10.1007/s11207-019-1517-4 fatcat:mhxms3fznjfbrhd3a2ydjvv3ky

Deep Learning-based Physical-Layer Secret Key Generation for FDD Systems [article]

Xinwei Zhang, Guyue Li, Junqing Zhang, Aiqun Hu, Zongyue Hou, Bin Xiao
2021 arXiv   pre-print
In particular, this is the first time to apply deep learning for PKG in FDD systems.  ...  This paper proposes a novel PKG scheme that uses the feature mapping function between different frequency bands obtained by deep learning to make two users generate highly similar channel features in FDD  ...  Deep Learning for Wireless Physical Layer Deep learning has been introduced to the wireless physical layer and achieved excellent performance in many areas such as channel estimation [18] , CSI feedback  ... 
arXiv:2105.08364v2 fatcat:cjculuntq5hjdoilhforohubzy

A deep learning-based reconstruction of cosmic ray-induced air showers

M. Erdmann, J. Glombitza, D. Walz
2018 Astroparticle physics  
We describe a method of reconstructing air showers induced by cosmic rays using deep learning techniques.  ...  We simulate an observatory consisting of ground-based particle detectors with fixed locations on a regular grid.  ...  For an overview of deep learning techniques see [7] .  ... 
doi:10.1016/j.astropartphys.2017.10.006 fatcat:m2xutyvlqbcbdpdnsxlkrqquni
« Previous Showing results 1 — 15 out of 464,215 results