Maximum Entropy Learning with Deep Belief Networks

Entropy

Payton Lin¹  Szu-Wei Fu¹  Ying-Hui Lai¹  Syu-Siang Wang¹  Yu Tsao¹ 

[1] Research Center for Information Technology Innovation, Academia Sinica, Taipei 11529, Taiwan;

关键词: maximum entropy;    machine learning;    deep learning;    deep belief networks;    restricted Boltzmann machine;    deep neural networks;    low-resource tasks;   

DOI  :  10.3390/e18070251

来源: DOAJ

【 摘 要 】

Conventionally, the maximum likelihood (ML) criterion is applied to train a deep belief network (DBN). We present a maximum entropy (ME) learning algorithm for DBNs, designed specifically to handle limited training data. Maximizing only the entropy of parameters in the DBN allows more effective generalization capability, less bias towards data distributions, and robustness to over-fitting compared to ML learning. Results of text classification and object recognition tasks demonstrate ME-trained DBN outperforms ML-trained DBN when training data is limited.

【 授权许可】

Unknown