Abstract | ||
---|---|---|
In this paper we propose a new distributed learn- ing method called distributed network boosting (DNB) algorithm for distributed applications. The learned hy- potheses are exchanged between neighboring sites dur- ing learning process. Theoretical analysis shows that the DNB algorithm minimizes the cost function through the collaborative functional gradient descent in hy- potheses space. Comparison results of the DNB algo- rithm with other distributed learning methods on real data sets with different sizes show its effectiveness. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1109/ICPR.2008.4761440 | IJPRAI |
Keywords | Field | DocType |
distributed algorithms,collaborative learning algorithm,cost function minimization,learning (artificial intelligence),pattern classification,distributed learning algorithm,classification problem,hypotheses space,gradient methods,minimisation,groupware,distributed network boosting algorithm,collaborative functional gradient descent,distributed environment,distributed databases,collaborative learning,algorithm design and analysis,gradient descent,collaboration,learning artificial intelligence,classification algorithms,cost function,distributed application,boosting | Gradient descent,Collaborative learning,Algorithm design,Computer science,Collaborative software,Distributed algorithm,Artificial intelligence,Boosting (machine learning),Distributed database,Statistical classification,Machine learning | Journal |
Volume | Issue | ISSN |
24 | 05 | 1051-4651 E-ISBN : 978-1-4244-2175-6 |
ISBN | Citations | PageRank |
978-1-4244-2175-6 | 0 | 0.34 |
References | Authors | |
16 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shijun Wang | 1 | 239 | 22.83 |
Changshui Zhang | 2 | 5506 | 323.40 |