Generalization performance of bipartite ranking algorithms with convex losses

This chapter gives a discussion of search processes on networks. It begins with a discussion of web search, including crawlers and web ranking algorithms such as PageRank. Search in distributed databases such as peer-to-peer networks is also discussed, including simple breadth-first search style algorithms and more advanced “supernode” approaches. Finally, network navigation is discussed at some length, motivated by consideration of Milgram's letter passing experiment. Kleinberg's variant of the small-world model is introduced and it is shown that efficient navigation is possible only for certain values of the model parameters. Similar results are also derived for the hierarchical model of Watts et al.

Download Full-text

Fault diagnosis of industrial robot reducer by an extreme learning machine with a level-based learning swarm optimizer

Advances in Mechanical Engineering ◽

10.1177/16878140211019540 ◽

2021 ◽

Vol 13 (5) ◽

pp. 168781402110195

Author(s):

Jianwen Guo ◽

Xiaoyan Li ◽

Zhenpeng Lao ◽

Yandong Luo ◽

Jiapeng Wu ◽

...

Keyword(s):

Fault Diagnosis ◽

Extreme Learning Machine ◽

Large Scale ◽

Production Efficiency ◽

Industrial Robot ◽

Optimal Solution ◽

Industrial Robots ◽

Generalization Performance ◽

Gradient Descent Algorithm ◽

Learning Machine

Fault diagnosis is of great significance to improve the production efficiency and accuracy of industrial robots. Compared with the traditional gradient descent algorithm, the extreme learning machine (ELM) has the advantage of fast computing speed, but the input weights and the hidden node biases that are obtained at random affects the accuracy and generalization performance of ELM. However, the level-based learning swarm optimizer algorithm (LLSO) can quickly and effectively find the global optimal solution of large-scale problems, and can be used to solve the optimal combination of large-scale input weights and hidden biases in ELM. This paper proposes an extreme learning machine with a level-based learning swarm optimizer (LLSO-ELM) for fault diagnosis of industrial robot RV reducer. The model is tested by combining the attitude data of reducer gear under different fault modes. Compared with ELM, the experimental results show that this method has good stability and generalization performance.

Download Full-text

Comparison of ranking algorithms with dataspace

2015 International Conference on Advances in Computer Engineering and Applications ◽

10.1109/icacea.2015.7164756 ◽

2015 ◽

Cited By ~ 5

Author(s):

Niranjan Lal ◽

Samimul Qamar

Keyword(s):

Ranking Algorithms

Download Full-text

A gain perturbation method to improve the generalization performance for the recurrent neural network misfire detector

IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339) ◽

10.1109/ijcnn.1999.832599 ◽

2003 ◽

Author(s):

P. Sun ◽

K. Marko

Keyword(s):

Neural Network ◽

Perturbation Method ◽

Recurrent Neural Network ◽

Generalization Performance

Download Full-text

Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study

PLoS Medicine ◽

10.1371/journal.pmed.1002683 ◽

2018 ◽

Vol 15 (11) ◽

pp. e1002683 ◽

Cited By ~ 171

Author(s):

John R. Zech ◽

Marcus A. Badgeley ◽

Manway Liu ◽

Anthony B. Costa ◽

Joseph J. Titano ◽

...

Keyword(s):

Deep Learning ◽

Learning Model ◽

Cross Sectional Study ◽

Chest Radiographs ◽

Sectional Study ◽

Cross Sectional ◽

Generalization Performance ◽

Deep Learning Model

Download Full-text

Informative Gene Selection and Direct Classification of Tumor Based on Chi-Square Test of Pairwise Gene Interactions

BioMed Research International ◽

10.1155/2014/589290 ◽

2014 ◽

Vol 2014 ◽

pp. 1-9 ◽

Cited By ~ 3

Author(s):

Hongyan Zhang ◽

Lanzhi Li ◽

Chao Luo ◽

Congwei Sun ◽

Yuan Chen ◽

...

Keyword(s):

Feature Selection ◽

Gene Selection ◽

Gene Interactions ◽

Selection Methods ◽

Chi Square ◽

Generalization Performance ◽

Independent Test ◽

Chi Square Test ◽

Leave One Out ◽

Tumor Gene

In efforts to discover disease mechanisms and improve clinical diagnosis of tumors, it is useful to mine profiles for informative genes with definite biological meanings and to build robust classifiers with high precision. In this study, we developed a new method for tumor-gene selection, the Chi-square test-based integrated rank gene and direct classifier (χ2-IRG-DC). First, we obtained the weighted integrated rank of gene importance from chi-square tests of single and pairwise gene interactions. Then, we sequentially introduced the ranked genes and removed redundant genes by using leave-one-out cross-validation of the chi-square test-based Direct Classifier (χ2-DC) within the training set to obtain informative genes. Finally, we determined the accuracy of independent test data by utilizing the genes obtained above withχ2-DC. Furthermore, we analyzed the robustness ofχ2-IRG-DC by comparing the generalization performance of different models, the efficiency of different feature-selection methods, and the accuracy of different classifiers. An independent test of ten multiclass tumor gene-expression datasets showed thatχ2-IRG-DC could efficiently control overfitting and had higher generalization performance. The informative genes selected byχ2-IRG-DC could dramatically improve the independent test precision of other classifiers; meanwhile, the informative genes selected by other feature selection methods also had good performance inχ2-DC.

Download Full-text

Information Minimization to Improve Generalization Performance

ICANN ’94 ◽

10.1007/978-1-4471-2097-1_169 ◽

1994 ◽

pp. 717-720

Author(s):

Ryotaro Kamimura ◽

Shohachiro Nakanishi

Keyword(s):

Generalization Performance

Download Full-text

Source identification of infectious diseases in networks via label ranking

PLoS ONE ◽

10.1371/journal.pone.0245344 ◽

2021 ◽

Vol 16 (1) ◽

pp. e0245344

Author(s):

Jianye Zhou ◽

Yuewen Jiang ◽

Biqing Huang

Keyword(s):

Infectious Diseases ◽

Real World ◽

Source Identification ◽

Human Society ◽

Practical Applications ◽

Ranking Algorithms ◽

Label Ranking ◽

Infection Source ◽

And Control ◽

Complete Observation

Background Outbreaks of infectious diseases would cause great losses to the human society. Source identification in networks has drawn considerable interest in order to understand and control the infectious disease propagation processes. Unsatisfactory accuracy and high time complexity are major obstacles to practical applications under various real-world situations for existing source identification algorithms. Methods This study attempts to measure the possibility for nodes to become the infection source through label ranking. A unified Label Ranking framework for source identification with complete observation and snapshot is proposed. Firstly, a basic label ranking algorithm with complete observation of the network considering both infected and uninfected nodes is designed. Our inferred infection source node with the highest label ranking tends to have more infected nodes surrounding it, which makes it likely to be in the center of infection subgraph and far from the uninfected frontier. A two-stage algorithm for source identification via semi-supervised learning and label ranking is further proposed to address the source identification issue with snapshot. Results Extensive experiments are conducted on both synthetic and real-world network datasets. It turns out that the proposed label ranking algorithms are capable of identifying the propagation source under different situations fairly accurately with acceptable computational complexity without knowing the underlying model of infection propagation. Conclusions The effectiveness and efficiency of the label ranking algorithms proposed in this study make them be of practical value for infection source identification.

Download Full-text