Abstract
When there are multiple component predictors, it is promising to integrate them into one predictor for advanced reasoning. If each component predictor is given as a stochastic model in the form of probability distribution, an exponential mixture of the component probability distributions provides a good way to integrate them. However, weight parameters used in the exponential mixture model are difficult to estimate if there is no data for performance evaluation. As a suboptimal way to solve this problem, weight parameters may be estimated so that the exponential mixture model should be a balance point that is defined as an equilibrium point with respect to the distance from/to all component probability distributions. In this paper, we propose a weight parameter estimation method that represents this concept using a symmetric Kullback-Leibler divergence and discuss the features of this method.
| Original language | English |
|---|---|
| Title of host publication | 2014 Joint 7th International Conference on Soft Computing and Intelligent Systems, SCIS 2014 and 15th International Symposium on Advanced Intelligent Systems, ISIS 2014 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 1126-1129 |
| Number of pages | 4 |
| ISBN (Electronic) | 9781479959556 |
| DOIs | |
| Publication status | Published - 2014 Feb 18 |
| Externally published | Yes |
| Event | 2014 Joint 7th International Conference on Soft Computing and Intelligent Systems, SCIS 2014 and 15th International Symposium on Advanced Intelligent Systems, ISIS 2014 - Kitakyushu, Japan Duration: 2014 Dec 3 → 2014 Dec 6 |
Other
| Other | 2014 Joint 7th International Conference on Soft Computing and Intelligent Systems, SCIS 2014 and 15th International Symposium on Advanced Intelligent Systems, ISIS 2014 |
|---|---|
| Country | Japan |
| City | Kitakyushu |
| Period | 14/12/3 → 14/12/6 |
Fingerprint
Keywords
- ensemble learning
- exponential mixture model
- parameter estimation
- symmetric Kullback-Leibler divergence
ASJC Scopus subject areas
- Software
- Artificial Intelligence
Cite this
Unsupervised weight parameter estimation for exponential mixture distribution based on symmetric Kullback-Leibler divergence. / Uchida, Masato.
2014 Joint 7th International Conference on Soft Computing and Intelligent Systems, SCIS 2014 and 15th International Symposium on Advanced Intelligent Systems, ISIS 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 1126-1129 7044722.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Unsupervised weight parameter estimation for exponential mixture distribution based on symmetric Kullback-Leibler divergence
AU - Uchida, Masato
PY - 2014/2/18
Y1 - 2014/2/18
N2 - When there are multiple component predictors, it is promising to integrate them into one predictor for advanced reasoning. If each component predictor is given as a stochastic model in the form of probability distribution, an exponential mixture of the component probability distributions provides a good way to integrate them. However, weight parameters used in the exponential mixture model are difficult to estimate if there is no data for performance evaluation. As a suboptimal way to solve this problem, weight parameters may be estimated so that the exponential mixture model should be a balance point that is defined as an equilibrium point with respect to the distance from/to all component probability distributions. In this paper, we propose a weight parameter estimation method that represents this concept using a symmetric Kullback-Leibler divergence and discuss the features of this method.
AB - When there are multiple component predictors, it is promising to integrate them into one predictor for advanced reasoning. If each component predictor is given as a stochastic model in the form of probability distribution, an exponential mixture of the component probability distributions provides a good way to integrate them. However, weight parameters used in the exponential mixture model are difficult to estimate if there is no data for performance evaluation. As a suboptimal way to solve this problem, weight parameters may be estimated so that the exponential mixture model should be a balance point that is defined as an equilibrium point with respect to the distance from/to all component probability distributions. In this paper, we propose a weight parameter estimation method that represents this concept using a symmetric Kullback-Leibler divergence and discuss the features of this method.
KW - ensemble learning
KW - exponential mixture model
KW - parameter estimation
KW - symmetric Kullback-Leibler divergence
UR - http://www.scopus.com/inward/record.url?scp=84988288801&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84988288801&partnerID=8YFLogxK
U2 - 10.1109/SCIS-ISIS.2014.7044722
DO - 10.1109/SCIS-ISIS.2014.7044722
M3 - Conference contribution
AN - SCOPUS:84988288801
SP - 1126
EP - 1129
BT - 2014 Joint 7th International Conference on Soft Computing and Intelligent Systems, SCIS 2014 and 15th International Symposium on Advanced Intelligent Systems, ISIS 2014
PB - Institute of Electrical and Electronics Engineers Inc.
ER -