All Results

A New Interpolation Approach and Corresponding Instance-Based Learning

廉师友Subjects: Computer Science >> Other Disciplines of Computer Science

Starting from finding approximate value of a function, introduces the measure of approximation-degree between two numerical values, proposes the concepts of "strict approximation" and "strict approximation region", then, derives the corresponding one-dimensional interpolation methods and formulas, and then presents a calculation model called "sum-times-difference formula" for high-dimensional interpolation, thus develops a new interpolation approach ? ADB interpolation. ADB interpolation is applied to the interpolation of actual functions with satisfactory results. Viewed from principle and effect, the interpolation approach is of novel idea, and has the advantages of simple calculation, stable accuracy, facilitating parallel processing, very suiting for high-dimensional interpolation, and easy to be extended to the interpolation of vector valued functions. Applying the approach to instance-based learning, a new instance-based learning method ? learning using ADB interpolation ? is obtained. The learning method is of unique technique, which has also the advantages of definite mathematical basis, implicit distance weights, avoiding misclassification, high efficiency, and wide range of applications, as well as being interpretable, etc. In principle, this method is a kind of learning by analogy, which and the deep learning that belongs to inductive learning can complement each other, and for some problems, the two can even have an effect of “different approaches but equal results” in big data and cloud computing environment. Thus, the learning using ADB interpolation can also be regarded as a kind of “wide learning” that is dual to deep learning. |

Subjects: Computer Science >> Other Disciplines of Computer Science

负荷预测是电网系统中很多应用的关键部分，具有重要作用。然而，由于电网负荷的非线性、时变性和不确定性，使得准确预测负荷具有一定的挑战。充分挖掘负荷序列的潜在特征是提升预测准确率的关键。本文认为在特征提取时应该充分利用负荷序列的位置信息、趋势性、周期性和时间信息，同时还应构建更深层次的神经网络框架进行特征挖掘。因此，本文提出了基于特征嵌入和Transformer框架的负荷预测模型，该模型由特征嵌入层，Transformer层和预测层组成。在特征嵌入层，模型首先对历史负荷的位置信息、趋势性、周期性和时间信息进行特征嵌入，然后再与天气信息进行融合，得到特征向量。Transformer层则接受历史序列的特征向量并挖掘序列的非线性时序依赖关系。预测层通过全连接网络实现负荷预测。从实验结果来看，本文模型的预测性能优于对比模型，体现了该模型的可行性和有效性。 |

Subjects: Computer Science >> Other Disciplines of Computer Science

Complex evidence theory has been applied to several fields due to its advantages in modeling and processing uncertain information. However,to measure the uncertainty of the complex mass function is still an open issue. The main contribution of this paper is to propose a complex-valued Deng entropy. The complex-valued Deng entropy can effectively measure the uncertainty of the mass function in the complex-valued framework. Meanwhile, the complex-valued Deng entropy is a generalization of the Deng entropy and Shannon entropy. That is, the complex-valued Deng entropy can degenerate to classical Deng entropy when the complex-valued mass function degenerates to a mass function in real space. In addition, the proposed complex-valued Deng entropy can also degenerates to Shannon entropy when the complex-valued mass function degenerates to a probability distribution in real space. Some numerical examples demonstrate the compatibility and effectiveness of the complex-valued Deng entropy. |

Subjects: Computer Science >> Other Disciplines of Computer Science

Complex-valued expression models have been widely used in the application of intelligent decision systems. However, there is a lack of entropy to measure the uncertain information of the complex-valued probability distribution. Therefore, how to reasonably measure the uncertain information of the complex-valued probability distribution is a gap to be filled. In this paper, inspired by the Renyi entropy, we propose the Complex-valued Renyi entropy, which can measure uncertain information of the complex-valued probability distribution under the framework of complex numbers, and is also the first time to measure uncertain information in the complex space. The Complex-valued Renyi entropy contains the features of the classical Renyi entropy, i.e., the Complex-valued Renyi Entropy corresponds to different information functions with different parameters q. Meanwhile, the Complex-valued Renyi entropy has some properties, such as non-negativity, monotonicity, etc. Some numerical examples can demonstrate the flexibilities and reasonableness of the Complex-valued Renyi entropy. |

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
Hits*13903*，
Downloads*2291*，
Comment
*0*

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, BingSubjects: 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
Hits*15205*，
Downloads*1069*，
Comment
*0*

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
Hits*11808*，
Downloads*1742*，
Comment
*0*

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
Hits*18406*，
Downloads*1550*，
Comment
*1*

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

TingTing Li; WeiYa Fan; YunSong LuoSubjects: 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
Hits*12905*，
Downloads*968*，
Comment
*0*

[1 Pages/ 9 Totals]