Causal structure-discovery techniques usually assume that all causes of more than one variable are observed. This is the so-called causal sufficiency assumption. In practice, it is untestable, and often violated. In t...
详细信息
ISBN:
(纸本)9781605609492
Causal structure-discovery techniques usually assume that all causes of more than one variable are observed. This is the so-called causal sufficiency assumption. In practice, it is untestable, and often violated. In this paper, we present an efficient causal structure-learning algorithm, suited for causally insufficient data. Similar to algorithms such as IC* and FCI, the proposed approach drops the causal sufficiency assumption and learns a structure that indicates (potential) latent causes for pairs of observed variables. Assuming a constant local density of the data-generating graph, our algorithm makes a quadratic number of conditional-independence tests w.r.t. the number of variables. We show with experiments that our algorithm is comparable to the state-of-the-art FCI algorithm in accuracy, while being several orders of magnitude faster on large problems. We conclude that MBCS* makes a new range of causally insufficient problems computationally tractable.
This paper introduces a new way for text-line extraction by integrating deep-learning based pre-classification and state-of-the-art segmentation methods. Text-line extraction in complex handwritten documents poses a s...
详细信息
This paper presents a novel computational optimization of the deceived non local means filter using moving average and symmetric weighting. The proposed optimization is compared with different approaches that reduce t...
详细信息
In this work we explore a set of image enhancement techniques for improving contrast and removing noise from digital images of cell activity. The cells studied were extracted from cancerous brain tissue and exposed to...
详细信息
This paper presents a novel computational optimization of the deceived non local means filter using moving average and symmetric weighting. The proposed optimization is compared with different approaches that reduce t...
详细信息
This paper presents a novel computational optimization of the deceived non local means filter using moving average and symmetric weighting. The proposed optimization is compared with different approaches that reduce the computational cost of the deceived non local means filter. Furthermore, the impact of parallelizing different optimization approaches is assessed by evaluating the execution time and scalability in Xeon Phi KNL architecture. The proposed optimization for the sequential implementation achieved a 90x speedup, while its parallelized implementation yielded a speedup of up to 1662x.
This paper presents a systematic literature review of image datasets for document image analysis, focusing on historical documents, such as handwritten manuscripts and early prints. Finding appropriate datasets for hi...
详细信息
Comparing two graphs is a computationally difficult task [9], [8]. After a work by E. Arias-Mendez and F. Torres-Rojas [7] about the correlation of metabolic pathways with two new proposed approaches to simplify the c...
详细信息
Comparing two graphs is a computationally difficult task [9], [8]. After a work by E. Arias-Mendez and F. Torres-Rojas [7] about the correlation of metabolic pathways with two new proposed approaches to simplify the comparison of its associated graph representation, we extended this work to general graph structures as a simple way to compare them. The approach presented here is an extension of those algorithms to general graphs. The first algorithm proposed looks to transform the comparing graphs into linear sequences, to be analyzed using sequence-alignment tools from bioinformatics and get a numeric score as its value of similitude. The second proposed algorithm consists of the search of equal connected nodes between 2 graphs to eliminate then on both structures, only leaving the differences, as heuristic for comparison. These algorithms were developed as a low-cost process to correlate metabolic pathways showing good results; the suggestion is to use this information as a previous analysis to a deeper, more expensive, comparing tools use. Here we review the extension of this work as an application to a more general graph data structure. These methods have shown to be an effective way to treat the problem as listed in the results section.
We present an anatomically guided feature selection scheme for prediction of neurological disorders based on brain connectivity networks. Using anatomical information not only gives rise to an interpretable model, but...
详细信息
This paper presents an automatic approach for optimal calibration of the deceived non local means filter (DNLM), for enhancing segmentation accuracy of fluorescence based microscopy images. The DNLM is designed for im...
详细信息
ISBN:
(纸本)9781538680599
This paper presents an automatic approach for optimal calibration of the deceived non local means filter (DNLM), for enhancing segmentation accuracy of fluorescence based microscopy images. The DNLM is designed for image denoising and enhancement. The calibration of its parameters in real image preprocessing applications is very often time consuming, since doing a manual calibration in a sample image might not work in different image samples. We compared three different stochastic optimization approaches: simulated annealing, particle swarm optimization and genetic algorithms, and selected the best approach. The implemented solution needs from the user only to define a precision metric and a set of image samples, and the algorithm will arrive to a locally optimal set of the filter parameters, to improve segmentation accuracy, using Otsu thresholding and measured with the Dice index. The PSO approach presented the overall best performance, with an average Dice index of 0.9667 in the validation set, a two percent boost over the best manually calibrated set of parameters for the DNLM.
This paper introduces a new way for text-line extraction by integrating deep-learning based pre-classification and state-of-the-art segmentation methods. Text-line extraction in complex handwritten documents poses a s...
详细信息
暂无评论