TY - GEN

T1 - Acceleration for both Boltzmann Machine Learning and Mean Field Theory Learning

AU - Hagiwara, Masafumi

N1 - Publisher Copyright:
© 1992 IEEE.

PY - 1992

Y1 - 1992

N2 - This paper proposes new learning algorithms for both the Boltzmann Machine (BM) learning and the Mean Field Theory (MFT) learning to accelerate their learning speeds. The derivation of the new algorithms are based on the following assumptions: 1) The alternative cost function is {equation presented} where G τis the information-theoretical measure at the learning epoch τ, not G which is the commonly used information-theoretical measure in the derivation of BM learning. 2) The most recent weights are assumed in calculating Gn, which technique is used in the derivation of Recursive Least-Squares (RLS) algorithm. As a result, momentum terms which accelerate learning can be derived in the BM and the MFT learning algorithms. Comparing the proposed MFT learning algorithm with the conventional MFT algorithm by computer simulation, we show the effectiveness of the proposed method.

AB - This paper proposes new learning algorithms for both the Boltzmann Machine (BM) learning and the Mean Field Theory (MFT) learning to accelerate their learning speeds. The derivation of the new algorithms are based on the following assumptions: 1) The alternative cost function is {equation presented} where G τis the information-theoretical measure at the learning epoch τ, not G which is the commonly used information-theoretical measure in the derivation of BM learning. 2) The most recent weights are assumed in calculating Gn, which technique is used in the derivation of Recursive Least-Squares (RLS) algorithm. As a result, momentum terms which accelerate learning can be derived in the BM and the MFT learning algorithms. Comparing the proposed MFT learning algorithm with the conventional MFT algorithm by computer simulation, we show the effectiveness of the proposed method.

UR - http://www.scopus.com/inward/record.url?scp=84968426654&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84968426654&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.1992.287107

DO - 10.1109/IJCNN.1992.287107

M3 - Conference contribution

AN - SCOPUS:84968426654

T3 - Proceedings of the International Joint Conference on Neural Networks

SP - 687

EP - 692

BT - Proceedings - 1992 International Joint Conference on Neural Networks, IJCNN 1992

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1992 International Joint Conference on Neural Networks, IJCNN 1992

Y2 - 7 June 1992 through 11 June 1992

ER -