Transfer Learning Algorithm Combined with Hierarchy Correlation of Data

Hua Pang; Long Wang

doi:10.54691/fse.v2i11.2973

Authors

Hua Pang
Long Wang

DOI:

https://doi.org/10.54691/fse.v2i11.2973

Keywords:

Transfer learning, Hierarchical Correlation, Weight Distribution, Weight Adjustment

Abstract

An improved transfer learning algorithm is proposed for data with hierarchical correlation which is introduced in the classical transfer learning algorithm TrAdaBoost. In the initial weight assignment phase, the hierarchical relationship between data is considered, and weight restriction are added. The experimental results show that the correct, precision and recall rate of the recommendation method based on the improved algorithm reach more than 85% on the learning resource sample data with hierarchical correlation.

Downloads

Download data is not yet available.

References

PAN S J, YANG QIANG. A Survey on Transfer Learning[J]. IEEE Trans on Knowledge and Data Engineering, 2010, 22(10): 1345 - 1359.

PARDOE D, STONE P. Boosting for regression transfer[A]. Piscataway:IEEEInc Press,2010:863-870.

S.J.Pan, V.W.Zheng, Q.Yang, and D.H.Hu. Transfer learning for wifi-based indoor localization[C], the Proceedings of the Workshop on Transfer Learning for Complex Task of the 23rd AAAI Conference on Artificial Intelligence, Chicago, Illinois, USA, July 2008.

J.Ramon, K.Driessens, and T.Croonenborghs. Transfer learning in reinforcement learning problems through partial policy recycling[C], the Proceedings of the 18th European conference on Machine Learning, Berlin, Heidelberg: Springer-Verlag, 2007, pp. 699–707.

M.E.Taylor and P.Stone. Cross-domain transfer for reinforcement learning[C], the Proceedings of the 24th international conference on Machine learning. New York, USA: ACM, 2007, pp. 879–886.

DAI WENYUAN, YANG Q, XUE G. Boosting for transfer learning[A]. New York:Academic Press,Inc, 2007:193-200.