A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf
.
Filters
Genesis of Basic and Multi-Layer Echo State Network Recurrent Autoencoders for Efficient Data Representations
[article]
2018
arXiv
pre-print
Echo State Network (ESN) is a recent specific kind of Recurrent Neural Network which presents very rich dynamics thanks to its reservoir-based hidden layer. ...
In order to bring up sturdier alternative to conventional reservoir-based networks, not only single layer basic ESN is used as an autoencoder, but also Multi-Layer ESN (ML-ESN-RAE). ...
ACKNOWLEDGMENT The research leading to these results has received funding from the Ministry of Higher Education and Scientific Research of Tunisia under the grant agreement number LR11ES48. ...
arXiv:1804.08996v2
fatcat:jl3io4v3bbdplnqmphcqgu6yfy
Hierarchical Bi-level Multi-Objective Evolution of Single- and Multi-layer Echo State Network Autoencoders for Data Representations
[article]
2018
arXiv
pre-print
ESN is not only used with its basic single layer form but also with the recently proposed Multi-Layer (ML) architecture. ...
Echo State Network (ESN) presents a distinguished kind of recurrent neural networks. It is built upon a sparse, random and large hidden infrastructure called reservoir. ...
References [1] The Nature of Statistical Learning Theory, Springer, 2000. ...
arXiv:1806.01016v2
fatcat:z3yfsvmp6jgbtfoemxquom73du
Text feature extraction based on deep learning: a review
2017
EURASIP Journal on Wireless Communications and Networking
Selection of text feature item is a basic and important matter for text mining and information retrieval. Traditional methods of feature extraction require handcrafted features. ...
Deep learning can automatically learn feature representation from big data, including millions of parameters. ...
Acknowledgements This work is supported by supported by the Fundamental Research Funds for the Central Universities (Grant No.18CX02019A). ...
doi:10.1186/s13638-017-0993-1
pmid:29263717
pmcid:PMC5732309
fatcat:bqyk3wddqbebdfeki72myn5p2y
Survey on Deep Fuzzy Systems in regression applications: a view on interpretability
[article]
2022
arXiv
pre-print
Regression problems have been more and more embraced by deep learning (DL) techniques. ...
Fuzzy logic systems (FLS) are inherently interpretable models, well known in the literature, capable of using nonlinear representations for complex systems through linguistic terms with membership degrees ...
Each RBM comprises a layer of visible units for data representation and a layer of hidden units for feature representation, learned by capturing higher-order correlations from the data. ...
arXiv:2209.04230v1
fatcat:mvwyqrhb3nerndnedlq4ccidku
Neural Computing
[article]
2021
arXiv
pre-print
major researchers and innovators in this field and thus, encouraging the readers to develop newer and more advanced techniques for the same. ...
By means of this chapter, the societal problems are discussed and various solutions are also given by means of the theories presented and researches done so far. ...
Echo State Machine Echo state network (ESN) or machine, created by (Jaeger, 2001) , is a type of recurrent neural network with thinly distributed connections within the hidden layer, made up of recurrent ...
arXiv:2107.02744v1
fatcat:kmfb6j3vcrby3mphgzwo6akho4
A Meta-learning Approach to Reservoir Computing: Time Series Prediction with Limited Data
[article]
2021
arXiv
pre-print
Recent research has established the effectiveness of machine learning for data-driven prediction of the future evolution of unknown dynamical systems, including chaotic systems. ...
We demonstrate our approach on a simple benchmark problem, where it beats the state of the art meta-learning techniques, as well as a challenging chaotic problem. ...
We thank Edward Ott, Brian Hunt, and Jiangying Zhou for useful comments and discussion. ...
arXiv:2110.03722v1
fatcat:tk36gfoodjap7i2z4ns5tw4l6u
Learning a repertoire of actions with deep neural networks
2014
4th International Conference on Development and Learning and on Epigenetic Robotics
Taking a handwriting task as an example, we apply the deep learning paradigm to build a network which uses a high-level representation of digits to generate sequences of commands, directly fed to a low-level ...
We address the problem of endowing a robot with the capability to learn a repertoire of actions using as little prior knowledge as possible. ...
Other techniques such as echo state networks and liquid state machines [14] , [15] also use recurrent hidden layers but with randomly chosen weights which remain untrained 1 . ...
doi:10.1109/devlrn.2014.6982986
dblp:conf/icdl-epirob/DroniouIS14
fatcat:hrknn7pzynevnmht73fn2d37kq
Autoencoding Neural Networks as Musical Audio Synthesizers
[article]
2020
arXiv
pre-print
A method for musical audio synthesis using autoencoding neural networks is proposed. The autoencoder is trained to compress and reconstruct magnitude short-time Fourier transform frames. ...
Our algorithm is light-weight when compared to current state-of-the-art audio-producing machine learning algorithms. ...
Learning Task Description In our work we train a multi-layer autoencoder to learn representations of musical audio. ...
arXiv:2004.13172v1
fatcat:3exekjf4jrehfhxqxzpglgf4wu
Deep Randomized Neural Networks
[article]
2021
arXiv
pre-print
For both, we focus specifically on recent results in the domain of deep randomized systems, and (for recurrent models) their application to structured domains. ...
Typical examples of such systems consist of multi-layered neural network architectures where the connections to the hidden layer(s) are left untrained after initialization. ...
In particular, we keep our focus on the ESN formalism, extended to the multi-layer setting by the Deep Echo State Network (DeepESN) model. ...
arXiv:2002.12287v2
fatcat:iy6r4bzka5bitjdugsfgbpioia
Deep Learning in Spatiotemporal Cardiac Imaging: A Review of Methodologies and Clinical Usability
2020
Computers in Biology and Medicine
This review aims to synthesize the most relevant deep learning methods and discuss their clinical usability in dynamic cardiac imaging using for example the complete spatiotemporal image information of ...
Interestingly, not a single one of the reviewed papers was classified as a "clinical level" study. ...
[47] suggested a novel method to simultaneously estimate motion and segment cardiac structures from CMR cine sequences by using a novel Siamese style multi-scale recurrent network for unsupervised motion ...
doi:10.1016/j.compbiomed.2020.104200
pmid:33421825
fatcat:ltxjpt6yhzgvdkifo4wo3ftveq
Deep learning in systems medicine
2020
Briefings in Bioinformatics
Being able to automatically extract relevant features needed for a given task from high-dimensional, heterogeneous data, deep learning (DL) holds great promise in this endeavour. ...
Systems medicine (SM) has emerged as a powerful tool for studying the human body at the systems level with the aim of improving our understanding, prevention and treatment of complex diseases. ...
Autoencoders Autoencoders [74] are a typical DL model designed to learn efficient data representation in an unsupervised fashion. ...
doi:10.1093/bib/bbaa237
pmid:33197934
pmcid:PMC8382976
fatcat:bjhlu5jaubci3lm4j3vxiofehu
Deep Learning Approaches to Aircraft Maintenance, Repair and Overhaul: A Review
2018
2018 21st International Conference on Intelligent Transportation Systems (ITSC)
Although deep learning in general is not yet largely exploited for aircraft health, from our search, we identified four main architectures employed to MRO, namely, Deep Autoencoders, Long Short-Term Memory ...
These methods assist, for instance, with determining appropriate actions for aircraft maintenance, repair and overhaul (MRO). ...
It learns hidden representation and generates new data Difficult to implement and optimise compared to other variants of autoencoder. ...
doi:10.1109/itsc.2018.8569502
dblp:conf/itsc/RengasamyMF18
fatcat:fdetlamp6vesngmpu7icccvqxe
Universal Time-Series Representation Learning: A Survey
[article]
2024
arXiv
pre-print
This survey first presents a novel taxonomy based on three fundamental elements in designing state-of-the-art universal representation learning methods for time series. ...
Learning representations by extracting and inferring valuable information from these time series is crucial for understanding the complex dynamics of particular phenomena and enabling informed decisions ...
This subsection overviews basic neural network architectures used as building blocks in state-of-the-art representation learning methods for time series.
Multi-Layer Perceptrons (MLP). ...
arXiv:2401.03717v1
fatcat:zy6pafj4vjac7fpj7umfxjju3y
Speech representation learning: Learning bidirectional encoders with single-view, multi-view, and multi-task methods
[article]
2023
arXiv
pre-print
Unlike most other works that focus on a single learning setting, this thesis studies multiple settings: supervised learning with auxiliary losses, unsupervised learning, semi-supervised learning, and multi-view ...
Supervised learning has been the most dominant approach for training deep neural networks for learning good sequential representations. ...
I also appreciate all the collaboration and friendship from TTIC colleagues and my research collaborators, I learned a lot from all of them. ...
arXiv:2308.00129v1
fatcat:6iezeebyvvauppipt242y2gx5i
A Comprehensive Survey of Machine Learning Applied to Radar Signal Processing
[article]
2020
arXiv
pre-print
This work is amply introduced by providing general elements of ML-based RSP and by stating the motivations behind them. ...
With the rapid development of machine learning (ML), especially deep learning, radar researchers have started integrating these new methods when solving RSP-related problems. ...
., classes) with the high-level feature representation learning by multi-hidden layers. ...
arXiv:2009.13702v1
fatcat:m6am73324zdwba736sn3vmph3i
« Previous
Showing results 1 — 15 out of 565 results