Topological deep learning algorithm with visual impression: Fid-Bau Portal

Fachinformationsdienst BAUdigital

A platform for research: civil engineering, architecture and urbanism

Topological deep learning algorithm with visual impression

Yang, Mengduo / Li, Fanzhang

We present in this paper a novel approach for training a topological deep neural network with visual impression. We show that by combing denoising auto-encoder model and contractive auto-encoder with Hessian regularization model, we can achieve a deterministic auto-encoder aiming for robustness to small variations of the input. We exploit the tangent propagation algorithm to show how our algorithm can capture the manifold structure of the visual impression and build a topological atlas of charts. Finally, we show that by using the learned features to initialize a deep network, we achieve superior classification with relatively smaller parameters than some other models.

Access

Check availability in my library

Order at Subito €

Page navigation

Document information

Export, share and cite

Topological deep learning algorithm with visual impression

Yang, Mengduo / Li, Fanzhang

We present in this paper a novel approach for training a topological deep neural network with visual impression. We show that by combing denoising auto-encoder model and contractive auto-encoder with Hessian regularization model, we can achieve a deterministic auto-encoder aiming for robustness to small variations of the input. We exploit the tangent propagation algorithm to show how our algorithm can capture the manifold structure of the visual impression and build a topological atlas of charts. Finally, we show that by using the learned features to initialize a deep network, we achieve superior classification with relatively smaller parameters than some other models.

Topological deep learning algorithm with visual impression

Yang, Mengduo / Li, Fanzhang

We present in this paper a novel approach for training a topological deep neural network with visual impression. We show that by combing denoising auto-encoder model and contractive auto-encoder with Hessian regularization model, we can achieve a deterministic auto-encoder aiming for robustness to small variations of the input. We exploit the tangent propagation algorithm to show how our algorithm can capture the manifold structure of the visual impression and build a topological atlas of charts. Finally, we show that by using the learned features to initialize a deep network, we achieve superior classification with relatively smaller parameters than some other models.

Access

Check availability in my library

Order at Subito €

Page navigation

Document information

Export, share and cite

Document information

Title:

Topological deep learning algorithm with visual impression

Contributors:

Yang, Mengduo (author) / Li, Fanzhang (author)

Published in:

2017 International Smart Cities Conference (ISC2) ; 1-4

Publication date:

2017-09-01

Size:

201101 byte

ISBN:

978-1-5386-2523-1 , 978-1-5386-2524-8

DOI:

https://doi.org/10.1109/ISC2.2017.8090854

Type of media:

Conference paper

Type of material:

Electronic Resource

Language:

English

Similar titles

Deep learning algorithm with visual impression

He, Funan / Yang, Mengduo / Li, Fanzhang | IEEE | 2017

Lie group impression for deep learning

Yang, Mengduo / Li, Fanzhang | IEEE | 2017

A Novel Evaluation Method of Visual Impression of Black Fabrics

Asano, C.M. / Asano, A. / Murakami, M. et al. | British Library Conference Proceedings | 2003

Audio-visual Mutual Effect of Spatial Impression in Architectural Space

Hiraoka, Chiharu / Terashima, Takane / Ishikawa, Ayumi et al. | DataCite | 2019

Impression and Reflection

Zhu, W. | British Library Online Contents | 1997