Filters








8,829 Hits in 3.8 sec

Learning Temporal Rules from Noisy Timeseries Data [article]

Karan Samel, Zelin Zhao, Binghong Chen, Shuang Li, Dharmashankar Subramanian, Irfan Essa, Le Song
2022 arXiv   pre-print
Such salient composite events are provided as labels in temporal datasets and most works optimize models to predict these composite event labels directly.  ...  This is done through efficiently searching through the combinatorial space of all temporal logic rules in an end-to-end differentiable manner.  ...  Combinatorial Inference Instead of setting thresholds, we directly optimize over the space of combinatorial rules to infer our composite event label y r .  ... 
arXiv:2202.05403v1 fatcat:2qlxrrvvyze4hkiav7ijqxyu74

Knowledge Graph Identification [chapter]

Jay Pujara, Hui Miao, Lise Getoor, William Cohen
2013 Lecture Notes in Computer Science  
The extractions form an extraction graph and we refer to the task of removing noise, inferring missing information, and determining which candidate facts should be included into a knowledge graph as knowledge  ...  MPE inference in MLNs requires optimizing over combinatorial assignments of Boolean truth values.  ...  Moreover, the combinatorial explosion of Boolean assignments to random variables makes inference and learning in MLNs intractable optimization problems.  ... 
doi:10.1007/978-3-642-41335-3_34 fatcat:x3i4cbm57ndp5hw44aaio4orke

Smooth curve extraction by mean field annealing

Laurent H�rault, Radu Horaud
1995 Annals of Mathematics and Artificial Intelligence  
In this paper we attack the figure-ground discrimination problem from a combinatorial optimization perspective.  ...  Moreover, this cost function fits the constraints of a interacting spin system which in turn is a well suited physical model to solve hard combinatorial optimization problems.  ...  The optimization itself is carried out by relaxation labelling. Finally, curve points are labelled "1" and noise points are labelled "0".  ... 
doi:10.1007/bf01530832 fatcat:emf6htp6aff5pmcknopbmf6oiy

Multiplexed Spectral Imaging of 120 Different Fluorescent Labels

Alex M. Valm, Rudolf Oldenbourg, Gary G. Borisy, James P Brody
2016 PLoS ONE  
However, the number of distinguishable fluorophores is still limited by the unavoidable decrease in signal to noise ratio when fluorescence signals are fractionated over multiple wavelength bins.  ...  We have applied our labeling and analysis strategy to identify microbes labeled by fluorescence in situ hybridization and here demonstrate the ability to distinguish 120 differently labeled microbes in  ...  The spectra were modeled with Poisson distributed shot noise and were unmixed against four possible fluorophores as shown in Fig 5A.  ... 
doi:10.1371/journal.pone.0158495 pmid:27391327 pmcid:PMC4938436 fatcat:km4vrx43c5b27icwuggbhohete

Label Inference Attacks from Log-loss Scores [article]

Abhinav Aggarwal, Shiva Prasad Kasiviswanathan, Zekun Xu, Oluwaseyi Feyisetan, Nathanael Teissier
2021 arXiv   pre-print
Additionally, we present label inference algorithms (attacks) that succeed even under addition of noise to the log-loss scores and under limited precision arithmetic.  ...  Surprisingly, we show that for any finite number of label classes, it is possible to accurately infer the labels of the dataset from the reported log-loss score of a single carefully constructed prediction  ...  We achieve this by safeguarding against the worst case magnitude of the noise that can be added for bounded noise distributions.  ... 
arXiv:2105.08266v2 fatcat:v7yy6ehy2rgbvk2git74ghipui

A continuum of transcriptional identities visualized by combinatorial fluorescent in situ hybridization

L. M. Jakt, S. Moriwaki, S. Nishikawa
2012 Development  
We show here that combinatorial labeling of RNA molecules with several fluorescent dyes extends the number of genes that can be targeted simultaneously beyond the number of fluorophores used.  ...  We have used combinatorial FISH and image analysis to measure the transcript densities of six genes using three fluorophores.  ...  (A) FISH against Fli1, Cdh5 and Gata2 using probe sets containing 48 probes all labeled with a single fluorescent dye. (B) FISH against Gata2, Tal1, Etv2 and Fli1 using combinatorial probe sets.  ... 
doi:10.1242/dev.086975 pmid:23175635 fatcat:qj3nmzk65zfrxhf72hrqe6ke5a

Toward accurate real-time marker labeling for live optical motion capture

Shihong Xia, Le Su, Xinyu Fei, Han Wang
2017 The Visual Computer  
The key idea is to formulate the problem in a combinatorial optimization framework.  ...  We demonstrate the power of our approach by capturing a wide range of human movements and achieve the state-of-the-art accuracy by comparing against alternative methods and commercial system like VICON  ...  We compare the resulting animation of our method against Vicon and alternative labeling methods: YLD, UGM, LH and CP.  ... 
doi:10.1007/s00371-017-1400-y fatcat:dxavheew5fh3tjpitq7scnk6je

Local Perturb-and-MAP for Structured Prediction [article]

Gedas Bertasius, Qiang Liu, Lorenzo Torresani, Jianbo Shi
2016 arXiv   pre-print
Perturb-and-MAP models provide a promising alternative to CRFs, but require global combinatorial optimization and hence they are usable only on specific models.  ...  Conditional random fields (CRFs) provide a powerful tool for structured prediction, but cast significant challenges in both the learning and inference steps.  ...  In the last column we Gumbel noise with 0.25 signal-to-noise ratio.  ... 
arXiv:1605.07686v2 fatcat:vgezpek7n5eerid26lkb7hx56u

Shapecollage: Occlusion-Aware, Example-Based Shape Interpretation [chapter]

Forrester Cole, Phillip Isola, William T. Freeman, Frédo Durand, Edward H. Adelson
2012 Lecture Notes in Computer Science  
We infer the best set of localized shape patches over a graph of keypoints at multiple scales to produce a discontinuous shape representation we term a shape collage.  ...  A major difficulty in applying an example-based approach to shape interpretation is the combinatorial explosion of shape possibilities that occur at occluding contours.  ...  Define b as the label set to fit against and a as the other set. Define w as the vector of weights corresponding to the relative areas ofÔ i .  ... 
doi:10.1007/978-3-642-33712-3_48 fatcat:6ljhag4h7jdvvl3sl77zsl7usq

GSAShrink: A Novel Iterative Approach for Wavelet-Based Image Denoising

Alexandre L.M. Levada, Alberto Tannús, Nelson D.A. Mascarenhas
2009 2009 XXII Brazilian Symposium on Computer Graphics and Image Processing  
To approximate the MAP estimator, we propose GSAShrink, a modified version of a known combinatorial optimization algorithm based on non-cooperative game theory (Game Strategy Approach, or GSA).  ...  One way to estimate x j,k is through Bayesian inference, by adopting a MAP approach. We propose a novel iterative method based on the combinatorial optimization algorithm GSA.  ...  ALGORITHMS FOR BAYESIAN INFERENCE Now that the wavelet denoising is stated as a Bayesian inference problem, algorithms for approximating the MAP estimator are required.  ... 
doi:10.1109/sibgrapi.2009.8 dblp:conf/sibgrapi/LevadaTM09 fatcat:wpw2eist5fgfpdbowfsexfzrf4

Probabilistic drug connectivity mapping

Juuso A Parkkinen, Samuel Kaski
2014 BMC Bioinformatics  
We infer the relevance for retrieval by data-driven probabilistic modeling of the drug responses, resulting in probabilistic connectivity mapping, and further consider the available cell lines as different  ...  Model inference is carried out with a variational approximation, using the R package CCAGFA available in CRAN [11] . Details of the inference are given in the Appendix.  ...  Differential expression was computed against the mean of the treatment measurements for each batch, instead of the biological controls, as suggested by Iskar et al. [14] .  ... 
doi:10.1186/1471-2105-15-113 pmid:24742351 pmcid:PMC4011783 fatcat:mbxzgdvc6jgx3l6i47jufpcvqi

PhISCS - A Combinatorial Approach for Sub-perfect Tumor Phylogeny Reconstruction via Integrative use of Single Cell and Bulk Sequencing Data [article]

Salem Malikic, Simone Ciccolella, Farid Rashidi Mehrabadi, Camir Ricketts, Md. Khaledur Rahman, Ehsan Haghshenas, Daniel Seidman, Faraz Hach, Iman Hajirasouliha, S. Cenk Sahinalp
2018 bioRxiv   pre-print
In order to address such limitations, we, for the first time, introduce a new combinatorial formulation that integrates single cell sequencing data with matching bulk sequencing data, with the objective  ...  Available computational methods for tumor phylogeny inference via SCS typically aim to identify the most likely perfect phylogeny tree satisfying infinite sites assumption (ISA).  ...  We were not able to compare against OncoNEM since it terminated with an error for most of the input matrices nor against ddClone which does not infer phylogeny.  ... 
doi:10.1101/376996 fatcat:fvpohlcyt5c6rnnkfs77tnj52a

Inferring perturbation profiles of cancer samples [article]

Martin Franz-Xaver Pirkl, Niko Beerenwinkel
2020 bioRxiv   pre-print
We combine genetic aberrations with gene expression data in a causal network derived across patients to infer unobserved perturbations.  ...  However, inferring perturbation profiles from molecular alterations is challenging due to error-prone molecular measurements and incomplete coverage of all possible molecular causes of gene perturbations  ...  Interestingly, the accuracy is also more robust against Gaussian noise, due to the large sample size for only eight P-genes. The other methods break down completely.  ... 
doi:10.1101/2020.12.10.419077 fatcat:nueejecqwfeonptppoj2jfx6be

Inferring perturbation profiles of cancer samples

Martin Pirkl, Niko Beerenwinkel
2021 Bioinformatics  
We combine genetic aberrations with gene expression data in a causal network derived across patients to infer unobserved perturbations.  ...  However, inferring perturbation profiles from molecular alterations is challenging due to error-prone molecular measurements and incomplete coverage of all possible molecular causes of gene perturbations  ...  Interestingly, the accuracy is also more robust against Gaussian noise, due to the large sample size for only eight P-genes. The other methods break down completely.  ... 
doi:10.1093/bioinformatics/btab113 pmid:33617647 fatcat:bprv7sk7nffu7cyxlooua2ismy

PhISCS: a combinatorial approach for subperfect tumor phylogeny reconstruction via integrative use of single-cell and bulk sequencing data

Salem Malikic, Farid Rashidi Mehrabadi, Simone Ciccolella, Md. Khaledur Rahman, Camir Ricketts, Ehsan Haghshenas, Daniel Seidman, Faraz Hach, Iman Hajirasouliha, S. Cenk Sahinalp
2019 Genome Research  
Available computational methods for tumor phylogeny inference via single-cell sequencing (SCS) data typically aim to identify the most likely perfect phylogeny tree satisfying the infinite sites assumption  ...  We then describe a combinatorial formulation to solve this problem which ensures that several lineage constraints imposed by the use of variant allele frequencies (VAFs, derived from bulk sequence data  ...  Combinatorial, in particular integer linear programming (ILP), formulations for phylogeny inference have been available in the literature for a while.  ... 
doi:10.1101/gr.234435.118 pmid:31628256 pmcid:PMC6836735 fatcat:xhkueoiutvawzg6huxxj7wibbu
« Previous Showing results 1 — 15 out of 8,829 results