Current Location:home > Browse

1. chinaXiv:202007.00047 [pdf]

鲁棒模式识别研究进展

张煦尧; 刘成林
Subjects: Computer Science >> Other Disciplines of Computer Science

目前诸多模式识别任务的识别精度获得不断提升,在一些任务上甚至超越了人的水平。单从识别精度的角度来看,模式识别似乎已经是一个被解决了的问题。然而,高精度的模式识别系统在实际应用中依旧会出现不稳定和不可靠的现象。因此,开放环境下的鲁棒性成为制约模式识别技术发展的新瓶颈。实际上,在大部分模式识别模型和算法背后蕴含着三个基础假设:封闭世界假设、独立同分布假设、以及大数据假设。这三个假设直接或间接影响了模式识别系统的鲁棒性,并且是造成机器智能和人类智能之间差异的主要原因。本文简要论述如何通过打破三个基础假设来提升模式识别系统的鲁棒性。

submitted time 2020-07-29 Hits3371Downloads425 Comment 0

2. chinaXiv:202006.00176 [pdf]

Automated Radiological Impression Generation for Plain Chest X-rays with End to End Deep Learning

Zhang, Shuai; Xin, Xiaoyan; Shen, Jingtao; Guo, Yachong; Wang, Yang; Yang, Xianfeng; Wang, Jun; Zhang, Jian; Zhang, Bing
Subjects: Computer Science >> Other Disciplines of Computer Science

The chest X-Ray (CXR) is the one of the most common clinical exam used to diagnose thoracic diseases and abnormalities. The volume of CXR scans generated daily in hospitals is huge. Therefore, an automated diagnosis system that is able to save the effort of doctors is of great value. At present, the applications of artificial intelligence in CXR diagnosis usually use pattern recognition to classify the scans. However, such methods rely on labeled databases. They are costly and usually have a high error rate. In this work, we built a database containing more than 12,000 CXR scans and radiological reports, and developed a model based on deep convolutional neural network and recurrent network with attention mechanism. The model learns features from the CXR scans and the associated raw radiological reports directly; no additional labeling required. The model provides automated recognition of given scans and generation of impression. The quality of the generated impression was evaluated with both the CIDEr scores and by radiologists as well. The CIDEr scores were found to be around 5.8 on average for the testing dataset. Further blind evaluation suggested a comparable performance against radiologists.

submitted time 2020-06-09 Hits6006Downloads309 Comment 0

3. chinaXiv:202004.00006 [pdf]

一种新的结合仿生学的人工神经网络模型评估研究.pdf

张锦; 舒炫煜; 黄昭彦; 易胜
Subjects: Computer Science >> Other Disciplines of Computer Science

人工神经网络的模型结构与功能分别朝着多样化、智能化趋势发展,但研究者仅从解决问题结果的优劣对模型进行评估是有所欠缺、过于片面的。因此在本文中提出从仿生学的角度构建评估人工神经网络仿生度的指标集,采用定性与定量的方式对模型的仿生度进行整体分析。在定性方面,对模型的神经元方程、网络结构、权重更新原理等方面进行比较分析;在定量方面,基于仿生的角度构建指标集即小世界特性、同步特性及混沌特性,对模型进行分析,分析结果表明,LeNet5模型及BP神经网络具备同步特性,但其与真实生物神经网络仍有一定的距离,而KIII模型在结构上具备一定的小世界特性,其网络内部也表现同步特性及混沌特性,与真实的生物神经网络更为接近。

submitted time 2020-03-29 Hits10291Downloads863 Comment 0

4. chinaXiv:202003.00048 [pdf]

自监督图像增强网络:仅需低照度图像进行训练

张雨; 遆晓光; 张斌; 王春晖
Subjects: Computer Science >> Other Disciplines of Computer Science

本文提出了一种基于深度学习的自监督低照度图像增强方法。受信息熵理论和Retinex模型的启发,我们提出了一种基于信息熵最大的Retinex模型。利用该模型,一个非常简单的网络可以将照度图和反射图分离开来,且仅用低照度图像就可以进行训练。为了实现自监督学习,我们在模型中引入了一个约束条件:反射图的最大值通道与低照度图像的最大值通道一致,且其熵最大。我们的模型非常简单,不依赖任何精心设计的数据集(即使是一张低照度图像也能完成网络的训练),网络仅需进行分钟级的训练即可实现图像增强。实验证明,该方法在处理速度和效果上均达到了当前最新水平。

submitted time 2020-03-06 Hits16880Downloads774 Comment 1

5. chinaXiv:201903.00220 [pdf]

A method on selecting reliable samples based on fuzziness in positive and unlabeled learning.pdf

TingTing Li; WeiYa Fan; YunSong Luo
Subjects: Computer Science >> Other Disciplines of Computer Science

Traditional semi-supervised learning uses only labeled instances to train a classifier and then this classifier is utilized to classify unlabeled instances, while sometimes there are only positive instances which are elements of the target concept are available in the labeled set. Our research in this paper the design of learning algorithms from positive and unlabeled instances only. Among all the semi-supervised positive and unlabeled learning methods, it is a fundamental step to extract useful information from unlabeled instances. In this paper, we design a novel framework to take advantage of valid information in unlabeled instances. In essence, this framework mainly includes that (1) selects reliable negative instances through the fuzziness of the instances; (2) chooses new positive instances based on the fuzziness of the instances to expand the initial positive set, and we named these new instances as reliable positive instances; (3) uses data editing technique to filter out noise points with high fuzziness. The effectiveness of the presented algorithm is verified by comparative experiments on UCI dataset.

submitted time 2019-03-17 Hits8990Downloads486 Comment 0

  [1 Pages/ 5 Totals]