Filters








151 Hits in 7.1 sec

Semi-Supervised Cross-Subject Emotion Recognition Based on Stacked Denoising Autoencoder Architecture Using a Fusion of Multi-Modal Physiological Signals

Junhai Luo, Yuxin Tian, Hang Yu, Yu Chen, Man Wu
2022 Entropy  
We use two deep architectures and another generative stacked semi-supervised architecture as references for comparison to test the method's practical effects.  ...  It turns out that the fusion data perform better than the two modalities. To take advantage of deep-learning algorithms, we augment the original data and feed it directly into our training model.  ...  The fusion of two different physiological modalities is performed on a data-level, feature-level, and decision-level.  ... 
doi:10.3390/e24050577 pmid:35626462 fatcat:htfbm336ynhafpxh6ntdffv7ky

Deep Learning: Methods and Applications

Li Deng
2014 Foundations and Trends® in Signal Processing  
It offers a principled approach to LM smoothing by incorporating the power-law distribution for natural language.  ...  The above discussion used two of the most common conditional distributions for the visible data in the RBM -Gaussian (for continuous-valued data) and binomial (for binary data).  ... 
doi:10.1561/2000000039 fatcat:vucffxhse5gfhgvt5zphgshjy4

A tutorial survey of architectures, algorithms, and applications for deep learning

Li Deng
2014 APSIPA Transactions on Signal and Information Processing  
the same lower-level concepts help to define higher-level ones.  ...  classification and for feature learning.  ...  It offers a principled approach to LM smoothing by incorporating the power-law distribution for natural language.  ... 
doi:10.1017/atsip.2013.9 fatcat:4l4uonhhcffkbfot2fztpfxo2e

A Review on Community Detection in Large Complex Networks from Conventional to Deep Learning Methods: A Call for the Use of Parallel Meta-Heuristic Algorithms

Mohammed Al-Andoli, Shing Chiang Tan, Wooi Ping Cheah, Sin Yin Tan
2021 IEEE Access  
These distributions can be tuned according to a power law with different exponents.  ...  [81] Degeneracy for graph embedding using autoencoder 1M G Moderate + [79] Multiple Autoencoders- based embedding and clustering 1M T, G M.  ... 
doi:10.1109/access.2021.3095335 fatcat:4zggvxofqvbcjbwylk7swc3c34

Understanding Consumer Preferences Through Latent Spaces / Entendendo as preferências do consumidor através dos espaços latentes

André Uratsuka Manoel, Gustavo Corrêa Mirapalheta, João Luiz Chela
2021 Brazilian Journal of Technology  
It permits embedding vectors use. These are vectors in a k-dimensional vector space representing increasingly advanced study objects learning models, forming entirely new basis areas.  ...  Two k-dimensional Latent Vector Spaces representing film characteristics and the corresponding user preferences were created using Google Tensorflow, Machine Learning techniques like SGD and Matrix Factorization  ...  We investigated a number of choices for P n (w) and found that the unigram distribution U(w) raised to the 3/ 4rd power (i.e., U(w) 3/ 4 /Z)o utperformed significantly the unigram and the uniform distributions  ... 
doi:10.38152/bjtv4n1-004 fatcat:3ohnhwb57rhrnhimtsddg6b6da

Unsupervised learning of anomalous diffusion data [article]

Gorka Muñoz-Gil, Guillem Guigó i Corominas, Maciej Lewenstein
2021 arXiv   pre-print
At last, we showcase the use of the method in experimental data and demonstrate its advantages for cases where supervised learning is not applicable.  ...  In this work, we explore the use of unsupervised methods in anomalous diffusion data. We show that the main diffusion characteristics can be learnt without the need of any labelling of the data.  ...  However, in a bigger picture, traces of CTRW such as a broad distribution of trapping times will arise.  ... 
arXiv:2108.03411v1 fatcat:4z7qirbm4zbz3f7y4xcou6ytja

A Survey on Variational Autoencoders from a Green AI Perspective

Andrea Asperti, Davide Evangelista, Elena Loli Piccolomini
2021 SN Computer Science  
the generation problem for high-dimensional data.  ...  The key insight of VAEs is to learn the latent distribution of data in such a way that new meaningful samples can be generated from it.  ...  Asperti, is developing the library for the computation of flops used in this article.  ... 
doi:10.1007/s42979-021-00702-9 fatcat:zhny7v5eybbhxd74fboopesu5e

Self-Supervised Speech Representation Learning: A Review [article]

Abdelrahman Mohamed, Hung-yi Lee, Lasse Borgholt, Jakob D. Havtorn, Joakim Edin, Christian Igel, Katrin Kirchhoff, Shang-Wen Li, Karen Livescu, Lars Maaløe, Tara N. Sainath, Shinji Watanabe
2022 arXiv   pre-print
Other approaches rely on multi-modal data for pre-training, mixing text or visual data streams with speech.  ...  It is likewise difficult to apply this to dialects and languages for which only limited labeled data is available.  ...  Although the precise relationship between data size and performance has not been quantified, we can assume that it follows a law of diminishing returns (or power law), similar to observations for most  ... 
arXiv:2205.10643v1 fatcat:w3gm53o4unhkjfkvi4a3d7a3ay

Deep Face Recognition: A Survey [article]

Mei Wang, Weihong Deng
2020 arXiv   pre-print
Deep learning applies multiple processing layers to learn representations of data with multiple levels of feature extraction.  ...  Then, we summarize and compare the commonly used databases for both model training and evaluation.  ...  However, there is always a distribution change or domain shift between two data domains that can degrade the performance on test data.  ... 
arXiv:1804.06655v9 fatcat:i2yxh7bf45attlfjd3akyc65me

Deep Latent-Variable Models for Text Generation [article]

Xiaoyu Shen
2022 arXiv   pre-print
Deep latent-variable models, by specifying the probabilistic distribution over an intermediate latent process, provide a potential way of addressing these problems while maintaining the expressive power  ...  Recently deep neural network-based end-to-end architectures have been widely adopted.  ...  There are only two requirement for applying this framewok: • The data distribution x must be continuous.  ... 
arXiv:2203.02055v1 fatcat:sq3upxl7xvfnhigoc7apszomwu

Construction Technology and Quality Control of Power and Electrical Engineering Based on Convolutional Neural Network

Lei Xiao, Jian Su
2021 Security and Communication Networks  
of small datasets in the construction of electric power and electrical engineering; in this way, the relevant data are analyzed; by controlling the quality of construction, the quality problem has been  ...  In the context of the Internet era, more and more parties have begun to store, process, and analyze data, but the accompanying question is whether people are reasonable about the data under the impact  ...  two-end power distribution system [22] . eir naming is based on their specific topological structure, as shown in Figure 6 .  ... 
doi:10.1155/2021/8964532 fatcat:o6mi34rp4zh6pg7bzhwvmacgii

A comprehensive review of artificial intelligence-based approaches for rolling element bearing PHM: shallow and deep learning

Moussa Hamadache, Joon Ha Jung, Jungho Park, Byeng D. Youn
2019 JMST Advances  
Data-driven approaches, as opposed to model-based approaches, are gaining in popularity due to the availability of low-cost sensors and big data.  ...  Finally, deep-learning approaches for fault detection, diagnosis, and prognosis for REB are comprehensively reviewed.  ...  DL-based PHM techniques seek to handle big data by modeling the high-level illustration behind the data.  ... 
doi:10.1007/s42791-019-0016-y fatcat:sb3armogsvdebmwxjgpzfxwgju

Chaos as an interpretable benchmark for forecasting and data-driven modelling [article]

William Gilpin
2021 arXiv   pre-print
The striking fractal geometry of strange attractors underscores the generative nature of chaos: like probability distributions, chaotic systems can be repeatedly measured to produce arbitrarily-detailed  ...  The data license has been included elsewhere in this appendix. The authors have full control of the data repository on GitHub, and will ensure its continued accessibility.  ...  Acknowledgments and Disclosure of Funding We thank Gautam Reddy, Samantha Petti, Brian Matejek, and Yasa Baig for helpful discussions and comments on the manuscript. W.  ... 
arXiv:2110.05266v1 fatcat:762cndr4a5bzzbfkablmkvbk7u

Computational compound screening of biomolecules and soft materials by molecular simulations [article]

Tristan Bereau
2020 arXiv   pre-print
An account of the state of the art for the implementation of an MD-based screening paradigm is described, including automated force-field parametrization, system preparation, and efficient sampling across  ...  Emphasis is placed on machine-learning methods to enable MD-based screening.  ...  The combination of the two approaches effectively appears to achieve features in line with the variational autoencoder: the data-driven learning of a smooth latent-space distribution, coupled to a generative  ... 
arXiv:2010.03298v3 fatcat:4vm3vraaxnh57hgkfazqikfota

On closures for reduced order models - A spectrum of first-principle to machine-learned avenues [article]

Shady E. Ahmed, Suraj Pawar, Omer San, Adil Rasheed, Traian Iliescu, Bernd R. Noack
2021 arXiv   pre-print
For over a century, reduced order models (ROMs) have been a fundamental discipline of theoretical fluid mechanics.  ...  , and machine learning have changed the standard ROM methodology over the last two decades.  ...  Lee and Carlberg 455 developed a framework that computes the lower-dimensional embedding using a convolutional autoencoder and enforces physical conservation laws by modeling the latent-dynamics as a solution  ... 
arXiv:2106.14954v2 fatcat:q6jzbxfjabc3vg3nsn24z4lbyy
« Previous Showing results 1 — 15 out of 151 results