Filters








124 Hits in 4.3 sec

Dimensionality Reduction on Multi-Dimensional Transfer Functions for Multi-Channel Volume Data Sets

Han Suk Kim, Jürgen P. Schulze, Angela C. Cone, Gina E. Sosinsky, Maryann E. Martone
2010 Information Visualization  
In this article, we use Isomap and Locally Linear Embedding as well as a traditional algorithm, Principle Component Analysis.  ...  of transfer functions at a manageable level, that is, a maximum of three dimensions, which can be displayed visually in a straightforward way.  ...  Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. This publication is based, in part, on work supported Kim  ... 
doi:10.1057/ivs.2010.6 pmid:21841914 pmcid:PMC3153355 fatcat:y4vpulpo6jcwvbptlerasuih7q

Multichannel transfer function with dimensionality reduction

Han Suk Kim, Jürgen P. Schulze, Angela C. Cone, Gina E. Sosinsky, Maryann E. Martone, Chaomei Chen, Jinah Park, Ming C. Hao, Pak C. Wong
2010 Visualization and Data Analysis 2010  
In this paper, we used Isomap as well as a traditional algorithm, Principle Component Analysis (PCA).  ...  Our new method provides a framework to combine multiple approaches and pushes the boundary of gradient-based transfer functions to multiple channels, while still keeping the dimensionality of transfer  ...  ACKNOWLEDGEMENT This publication was made possible by Grant Number (NCRR P41-RR004050) from the National Center for Research Resources (NCRR), a part of the National Institutes of Health (NIH).  ... 
doi:10.1117/12.839526 pmid:20582228 pmcid:PMC2891081 dblp:conf/vda/KimSCSM10 fatcat:p5fwhnbfg5hhjibhq5ctw3brnu

Efficient cortical coding of 3D posture in freely behaving rats

Bartul Mimica, Benjamin A. Dunn, Tuce Tombaz, V. P. T. N. C. Srikanth Bojja, Jonathan R. Whitlock
2018 Science  
Both regions showed strong tuning to posture of the head, neck, and back, but signals for movement were much less dominant.  ...  Head and back representations were organized topographically across the PPC and M2, and more neurons represented postures that occurred less often.  ...  We therefore tracked the heads and backs of 11 rats in three dimensions while recording neural ensembles with dual microdrives targeting deep (>500 mm) layers of the PPC and M2, which exhibit thalamic,  ... 
doi:10.1126/science.aau2013 pmid:30385578 fatcat:2dtd56wpn5dtnifjrs6bxud7ni

Nonlinear Dimensionality Reduction Methods in Climate Data Analysis [article]

Ian Ross
2009 arXiv   pre-print
Linear dimensionality reduction techniques, notably principal component analysis, are widely used in climate data analysis as a means to aid in the interpretation of datasets of high dimensionality.  ...  The three methods used here are a nonlinear principal component analysis (NLPCA) approach based on neural networks, the Isomap isometric mapping algorithm, and Hessian locally linear embedding.  ...  This quicker convergence can be ascribed to better representation of the nonlinear ENSO variability by Isomap than by PCA.  ... 
arXiv:0901.0537v1 fatcat:2ethc7ddtjdyxkbqzmtg64upwi

Hyper-Path-Based Representation Learning for Hyper-Networks

Jie Huang, Xin Liu, Yangqiu Song
2019 Proceedings of the 28th ACM International Conference on Information and Knowledge Management - CIKM '19  
Network representation learning has aroused widespread interests in recent years.  ...  Finally, we conduct extensive experiments on several real-world datasets covering the tasks of link prediction and hyper-network reconstruction, and results demonstrate the rationality, validity, and effectiveness  ...  Basically, the 1D ConvNet takes the vector representations of the nodes in the sequence as input and outputs a latent vector representation of the sequence.  ... 
doi:10.1145/3357384.3357871 dblp:conf/cikm/00090S19 fatcat:ozle2gfc3nhpxedil3oednbo5a

Exemplar-Based EM-like image denoising via manifold reconstruction

Xin Li
2010 2010 IEEE International Conference on Image Processing  
Index Termsmanifold reconstruction, concentration of measure, blessing of dimensionality, EM-like iteration, image denoising.  ...  The byproduct of such manifold reconstruction from noisy data is an exemplar-Based EM-like (EBEM) denoising algorithm with minimal number of control parameters.  ...  Even with This work is in memorial of Sam Roweis whose ideas have deep influence on my research.  ... 
doi:10.1109/icip.2010.5652529 dblp:conf/icip/Li10 fatcat:v3uhcsdmabcxbkkii3wgmjbcgi

Topological Autoencoders [article]

Michael Moor and Max Horn and Bastian Rieck and Karsten Borgwardt
2021 arXiv   pre-print
We propose a novel approach for preserving topological structures of the input space in latent representations of autoencoders.  ...  We show that our approach is theoretically well-founded and that it exhibits favourable latent representations on a synthetic manifold as well as on real-world image data sets, while preserving low reconstruction  ...  representations with deep neural networks.  ... 
arXiv:1906.00722v5 fatcat:xiepjw3ji5hnxhkv55fmmz6y4i

Unsupervised Learning of Goal Spaces for Intrinsically Motivated Goal Exploration [article]

Alexandre Péré, Sébastien Forestier, Olivier Sigaud, Pierre-Yves Oudeyer
2018 arXiv   pre-print
In this work, we propose to use deep representation learning algorithms to learn an adequate goal space.  ...  representations.  ...  2D position of the ball in [0, 1] 2 , and for Arm-Arrow this is the 2D position and the 1D orientation of the arrow in [0, 1] 3 .  ... 
arXiv:1803.00781v3 fatcat:wlh5evfi5rdt3omwrwv6kpblda

A survey of unsupervised learning methods for high-dimensional uncertainty quantification in black-box-type problems [article]

Katiana Kontolati, Dimitrios Loukrezis, Dimitris G. Giovanis, Lohit Vandanapu, Michael D. Shields
2022 arXiv   pre-print
proposed in the literature, including recently proposed expensive deep neural network-based surrogates and can be readily applied for high-dimensional UQ in stochastic PDEs.  ...  complexity modeled as both Gaussian and non-Gaussian random fields to investigate the effect of the intrinsic dimensionality of input data.  ...  Another class of methods involve the use of Deep Neural Networks (DNNs) which provide a powerful tool to learn the latent input representation automatically by supervising model response.  ... 
arXiv:2202.04648v2 fatcat:xur2ilwqire6xjawbqarc2kv4e

A Tutorial on Graph Theory for Brain Signal Analysis [article]

Nikolaos Laskaris, Dimitrios A. Adamos, Anastasios Bezerianos
2020 arXiv   pre-print
Finally, the notion of signals residing on a given graph is introduced and elements from the emerging field of graph signal processing (GSP) are provided.  ...  The paper ends with a brief outline of the most recent trends in graph theory that are about to shape brain signal processing in the near future and a more general discussion on the relevance of graph-theoretic  ...  Figure 10 : 10 a) ISOMAP representation of attentive responses. b) MST-based representation of the trialto-trial response variability at PO7 sensor.  ... 
arXiv:2007.05800v1 fatcat:kg5dji4zxvaxfgn45uz6b3hhou

Deep Manifold Prior [article]

Matheus Gadelha, Rui Wang, Subhransu Maji
2020 arXiv   pre-print
We present a prior for manifold structured data, such as surfaces of 3D shapes, where deep neural networks are adopted to reconstruct a target shape using gradient descent starting from a random initialization  ...  reconstruction results on standard image to shape reconstruction benchmarks.  ...  ., IsoMap [35] or LLE [28] ). Our approach parameterizes the inverse mapping from the Euclidean space to the data manifold using a deep network.  ... 
arXiv:2004.04242v1 fatcat:lfqz7abwyfb7jo3uh2mz6m4sse

Autoencoder networks extract latent variables and encode these variables in their connectomes [article]

Matthew Farrell, Stefano Recanatesi, R. Clay Reid, Stefan Mihalas, Eric Shea-Brown
2020 bioRxiv   pre-print
Here, we define a tractable setting in which the problem of inferring circuit function from circuit connectivity can be analyzed in detail: the function of input compression and reconstruction, in an autoencoder  ...  Spectacular advances in imaging and data processing techniques are revealing a wealth of information about brain connectomes.  ...  ESB acknowledges the support of NSF DMS Grant 1514743. We thank the Allen Institute for Brain Science founders, Paul and Jody Allen, for their vision, encouragement, and support.  ... 
doi:10.1101/2020.03.04.977702 fatcat:hclefmow4zbsrkxs37wow3cbxa

Characterizing tissue composition through combined analysis of single-cell morphologies and transcriptional states [article]

Feng Bao, Yue Deng, Sen Wan, Bo Wang, Qionghai Dai, Steven Altschuler, Lani Wu
2020 bioRxiv   pre-print
Advances in spatial transcriptomics technologies enable optical profiling of morphological and transcriptional modalities from the same cells within tissues.  ...  MUSE provides a framework for combining multi-modal single-cell data to reveal deeper insights into the states, functions and organization of cells in complex biological tissues.  ...  representations into reconstructed features < % and < % .  ... 
doi:10.1101/2020.09.05.284539 fatcat:cqilt6z6cbfjfdhifytfdqf32a

The Structure Transfer Machine Theory and Applications [article]

Baochang Zhang, Lian Zhuo, Ze Wang, Jungong Han, Xiantong Zhen
2019 arXiv   pre-print
The resulting structure regularization term, named manifold loss, is incorporated into the loss function of the typical deep learning pipeline.  ...  Representation learning is a fundamental but challenging problem, especially when the distribution of data is unknown.  ...  In this paper, we theoretically show that the expected representation can be achieved in a probabilistic way as long as the property of manifold structure is revealed in the objective function of deep  ... 
arXiv:1804.00243v2 fatcat:dmlnobq75rf3jomd5xxrr7kkgu

Deep manifold learning reveals hidden dynamics of proteasome autoregulation [article]

Zhaolong Wu, Shuwen Zhang, Wei Li Wang, Yinping Ma, Yuanchen Dong, Youdong Mao
2021 arXiv   pre-print
Here we introduce a deep manifold learning framework, named AlphaCryo4D, which enables atomic-level cryogenic electron microscopy (cryo-EM) reconstructions of nonequilibrium conformational continuum and  ...  AlphaCryo4D integrates 3D deep residual learning with manifold embedding of free-energy landscapes, which directs 3D clustering via an energy-based particle-voting algorithm.  ...  ( Fig. 1d-f , Extended Data Fig. 2 , Methods).  ... 
arXiv:2012.12854v2 fatcat:g3xx3xzfinbutgmkwcrjkyikga
« Previous Showing results 1 — 15 out of 124 results