Filters








42,862 Hits in 2.4 sec

Graph visualization with latent variable models

Juuso Parkkinen, Kristian Nybo, Jaakko Peltonen, Samuel Kaski
2010 Proceedings of the Eighth Workshop on Mining and Learning with Graphs - MLG '10  
Then visualize similarity of the graph nodes with a suitable multidimensional scaling method, with similarity given by the model; we use a multidimensional scaling method optimized for a rigorous visual  ...  We suggest a novel modeling-driven approach to graph visualization: As usually in modeling, choose the (generative) model such that it captures what is important in the data.  ...  Generative Model for Graphs We assume that the links have been generated from a latent variable model, where the latent variables capture what is central in the graph, and the rest is noise.  ... 
doi:10.1145/1830252.1830265 dblp:conf/mlg/ParkkinenNPK10 fatcat:xb6jiqjcobfchfymuworsh6hc4

Disentangling Interpretable Generative Parameters of Random and Real-World Graphs [article]

Niklas Stoehr, Emine Yilmaz, Marc Brockschmidt, Jan Stuehmer
2019 arXiv   pre-print
When training our Beta-VAE model on ER random graphs, its latent variables have a near one-to-one mapping to the ER random graph parameters n and p.  ...  Our model is capable of learning disentangled, interpretable latent variables that represent the generative parameters of procedurally generated random graphs and real-world graphs.  ...  We generate a new set of 1,000 ER graphs with varying v k and feed these graphs to the trained model.  ... 
arXiv:1910.05639v2 fatcat:ruax6nzasrdrjcxuso25zgawqq

Visualization of Pairwise and Multilocus Linkage Disequilibrium Structure Using Latent Forests

Raphaël Mourad, Christine Sinoquet, Christian Dina, Philippe Leray, Konrad Scheffler
2011 PLoS ONE  
In this paper, we propose a method for short-range, long-range and chromosome-wide linkage disequilibrium visualization using forests of hierarchical latent class models.  ...  Probabilistic graphical models have been widely recognized as a powerful formalism allowing a concise and accurate modeling of dependences between variables.  ...  Latent Class models), able to cope with 10 5 variables and 2000 individuals [22, 23] .  ... 
doi:10.1371/journal.pone.0027320 pmid:22174739 pmcid:PMC3236755 fatcat:z4sxktipmjh2vcre5tp6r75rje

Weighted Network Density Predicts Range of Latent Variable Model Accuracy [article]

Jeremiah B. Palmerston, Qi She, Rosa Chan
2018 bioRxiv   pre-print
In this paper, a biologically plausible latent variable model was first fit to neural activity recorded via 2-photon microscopic calcium imaging in the murine primary visual cortex.  ...  Yet, relationships between these latent variable models and widely-studied network connectivity measures remained unclear.  ...  Recent research on latent variable models provides the mathematical techniques for such models with limited assumptions [7] .  ... 
doi:10.1101/343285 fatcat:7whx6snzc5cvrkwzn2lzosotky

Representation Learning for Dynamic Functional Connectivities via Variational Dynamic Graph Latent Variable Models

Yicong Huang, Zhuliang Yu
2022 Entropy  
In this work, we introduce a dynamic graph as the latent variable and develop a Variational Dynamic Graph Latent Variable Model (VDGLVM), a representation learning model based on the variational information  ...  The proposed computational model provides guaranteed behavior-decoding performance and improves LVMs by associating the inferred latent dynamics with probable DFC.  ...  This illustrates that using a dynamic graph as the latent variable does associate latent dynamics with probable DFC.  ... 
doi:10.3390/e24020152 pmid:35205448 pmcid:PMC8871213 fatcat:hfzi7hplnzboffchnf3lyrnxgq

A novel unsupervised analysis of electrophysiological signals reveals new sleep substages in mice

Vasiliki-Maria Katsageorgiou, Diego Sona, Matteo Zanotto, Glenda Lassi, Celina Garcia-Garcia, Valter Tucci, Vittorio Murino, Karunesh Ganguly
2018 PLoS Biology  
Moreover, the heterogeneous nature of sleep, with all its physiological aspects, is not fully accounted for by the current system of sleep stage classification.  ...  Sleep science is entering a new era, thanks to new data-driven analysis approaches that, combined with mouse gene-editing technologies, show a promise in functional genomics and translational research.  ...  Acknowledgments We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan X GPU used for the training of the mcRBM.  ... 
doi:10.1371/journal.pbio.2003663 pmid:29813050 pmcid:PMC5993302 fatcat:xtgzderltbeurldctwtrp5xg44

Categorical Normalizing Flows via Continuous Transformations [article]

Phillip Lippe, Efstratios Gavves
2021 arXiv   pre-print
Based on Categorical Normalizing Flows, we propose GraphCNF a permutation-invariant generative model on graphs.  ...  Instead, categorical data have complex and latent relations that must be inferred, like the synonymy between words.  ...  Graph theory with applications, volume 290. Macmillan London.  ... 
arXiv:2006.09790v3 fatcat:dwecz4ikjzhppirfzay57xy2cq

Fitting a Mixture Rasch Model to Visual Sequential Processing Memory Sub-dimension of ASIS: The Role of Covariates

Murat Doğan Şahin
2021 International Electronic Journal of Elementary Education  
With IRT, ability prediction independent of items and item parameter prediction independent of groups are ensured (Embretson & Reise, 2000; Hambleton & Swaminathan, 1985) .  ...  ) model which was also formed with three latent classes.  ...  However, a comparison between the two models could be carried out with this test because the first model with three latent classes in which covariate variables were not used was nested in the second (concomitant  ... 
doi:10.26822/iejee.2021.190 fatcat:3e4fwyaoebcdfdtbmxonnidu6a

Dimensionality Reduction for Data Visualization [Applications Corner]

Samuel Kaski, Jaakko Peltonen
2011 IEEE Signal Processing Magazine  
A promising direction forward is to learn a probabilistic latent variable model of the graph, in the hope of capturing its central properties, and then focus on visualizing those properties.  ...  For this experiment data, relevance is defined by the same data-driven biological processes being active, as modeled by a latent variable model (component model).  ... 
doi:10.1109/msp.2010.940003 fatcat:2rufhzzxpjevbc6zf3pbuykize

Bayesian Embedding of Co-occurrence Data for Query-Based Visualization

Mohammad Khoshneshin, W. Nick Street, Padmini Srinivasan
2011 2011 10th International Conference on Machine Learning and Applications and Workshops  
We propose a Bayesian approach to infer the latent variables. Given the intractability of inference for the posterior distribution, we use approximate inference via variational approaches.  ...  Our experiments show that our proposed models outperform co-occurrence data embedding, the state-of-the-art model for visualizing co-occurrence data.  ...  Let X i (a 1 × D vector) represent the latent variable of entity i and Y k (a 1 × D vector) represent the latent variable of entity k in a D-dimensional space.  ... 
doi:10.1109/icmla.2011.42 dblp:conf/icmla/KhoshneshinSS11 fatcat:7vptagrwtndmvkr6sdv4baq5ki

Weakly Correlated Knowledge Integration for Few-shot Image Classification

Chun Yang, Chang Liu, Xu-Cheng Yin
2022 Machine Intelligence Research  
To avoid explicitly aligning the visual features to the potentially biased and weakly correlated knowledge space, we sample a task-specific subgraph from UKG and append it as latent variables.  ...  Moreover, a graph attention module is proposed to sample the subgraph from the UKG with low complexity.  ...  The latent subgraph, which is an optimal subgraph of Gk with regard to task t. Used as a latent variable in the framework.  ... 
doi:10.1007/s11633-022-1320-9 fatcat:2r4igsbrk5ep3bccxsroexlrfm

Generating Tertiary Protein Structures via an Interpretative Variational Autoencoder [article]

Xiaojie Guo, Yuanqi Du, Sivani Tadepalli, Liang Zhao, Amarda Shehu
2021 arXiv   pre-print
Though typically deep generative models struggle with highly-structured data, the work presented here circumvents this challenge via graph-generative models.  ...  axes/factors that carry structural meaning and open the black box often associated with deep models.  ...  variable of latent code with other noise fixed.  ... 
arXiv:2004.07119v2 fatcat:6idinceogjdr5gvfpqmbonkn6e

Latent representation learning in biology and translational medicine

Andreas Kopf, Manfred Claassen
2021 Patterns  
Latent variable modeling allows for such interpretation by learning non-measurable hidden variables from observations.  ...  This review gives an overview over the different formal approaches to latent variable modeling, as well as applications at different scales of biological systems, such as molecular structures, intra- and  ...  and health. 1 Such variables are modeled as latent variables of a LVM.  ... 
doi:10.1016/j.patter.2021.100198 pmid:33748792 pmcid:PMC7961186 fatcat:d6ttueb5rbhotbsztha3wyvjt4

Fast Approximate Geodesics for Deep Generative Models [article]

Nutan Chen, Francesco Ferroni, Alexej Klushyn, Alexandros Paraschos, Justin Bayer, Patrick van der Smagt
2019 arXiv   pre-print
Our approach, therefore, is hence applicable to high-dimensional problems, e.g., in the visual domain.  ...  Current approaches are limited to low-dimensional latent spaces, due to the computational complexity of solving a non-convex optimisation problem.  ...  We will see that spanning the latent space of a latent variable model with a discrete and finite graph allows us to apply a classic search algorithm, A .  ... 
arXiv:1812.08284v2 fatcat:g7idww7i2rcgllrdmlinpnk6ru

Coney: a conversational approach to enhance engagement in surveys [article]

Damiano Scandolari, Mario Scrocca, Gloria Re Calegari, Irene Celino
2019 Zenodo  
Coney adopts a quantitative research method: survey questions are internally associated with a set of latent variables and each possible answer option is internally coded to allow for the numerical interpretation  ...  Coney relies on a graph-based data model for surveys.  ...  Given the possibility to tag questions with the investigated latent variable and to tag answers with their numerical coding, CONEY Inspect facilitates the basic analysis on collected data (e.g., frequency  ... 
doi:10.5281/zenodo.5846756 fatcat:bjbkrpyzyjbyfb6moxsp43nzl4
« Previous Showing results 1 — 15 out of 42,862 results