Abstract
There has been recent interest, and significant success, in adapting and extending ideas from statistical learning via Gaussian process (GP) regression to optimization via simulation (OvS) problems. At the heart of all such methods is a GP representing knowledge about the objective function whose conditional distribution is updated as more of the feasible region is explored. Calculating the conditional distribution requires inverting a large, dense covariance matrix, and this is the primary bottleneck for applying GP learning to large-scale OvS problems. If the GP is a Gaussian Markov Random Field (GMRF), then the precision matrix (inverse of the covariance matrix) can be constructed to be sparse. In this paper we show how to exploit this sparse-matrix structure to extend the reach of OvS based on GMRF learning for discrete-decision-variable problems.
| Original language | English (US) |
|---|---|
| Title of host publication | 2017 Winter Simulation Conference, WSC 2017 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 2080-2091 |
| Number of pages | 12 |
| ISBN (Electronic) | 9781538634288 |
| DOIs | |
| State | Published - Jan 4 2018 |
| Event | 2017 Winter Simulation Conference, WSC 2017 - Las Vegas, United States Duration: Dec 3 2017 → Dec 6 2017 |
Other
| Other | 2017 Winter Simulation Conference, WSC 2017 |
|---|---|
| Country | United States |
| City | Las Vegas |
| Period | 12/3/17 → 12/6/17 |
Fingerprint
ASJC Scopus subject areas
- Software
- Modeling and Simulation
- Computer Science Applications
Cite this
}
Computational methods for optimization via simulation using Gaussian Markov Random Fields. / Semelhago, Mark; Nelson, Barry L; Waechter, Andreas; Song, Eunhye.
2017 Winter Simulation Conference, WSC 2017. Institute of Electrical and Electronics Engineers Inc., 2018. p. 2080-2091.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - Computational methods for optimization via simulation using Gaussian Markov Random Fields
AU - Semelhago, Mark
AU - Nelson, Barry L
AU - Waechter, Andreas
AU - Song, Eunhye
PY - 2018/1/4
Y1 - 2018/1/4
N2 - There has been recent interest, and significant success, in adapting and extending ideas from statistical learning via Gaussian process (GP) regression to optimization via simulation (OvS) problems. At the heart of all such methods is a GP representing knowledge about the objective function whose conditional distribution is updated as more of the feasible region is explored. Calculating the conditional distribution requires inverting a large, dense covariance matrix, and this is the primary bottleneck for applying GP learning to large-scale OvS problems. If the GP is a Gaussian Markov Random Field (GMRF), then the precision matrix (inverse of the covariance matrix) can be constructed to be sparse. In this paper we show how to exploit this sparse-matrix structure to extend the reach of OvS based on GMRF learning for discrete-decision-variable problems.
AB - There has been recent interest, and significant success, in adapting and extending ideas from statistical learning via Gaussian process (GP) regression to optimization via simulation (OvS) problems. At the heart of all such methods is a GP representing knowledge about the objective function whose conditional distribution is updated as more of the feasible region is explored. Calculating the conditional distribution requires inverting a large, dense covariance matrix, and this is the primary bottleneck for applying GP learning to large-scale OvS problems. If the GP is a Gaussian Markov Random Field (GMRF), then the precision matrix (inverse of the covariance matrix) can be constructed to be sparse. In this paper we show how to exploit this sparse-matrix structure to extend the reach of OvS based on GMRF learning for discrete-decision-variable problems.
UR - http://www.scopus.com/inward/record.url?scp=85044522936&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85044522936&partnerID=8YFLogxK
U2 - 10.1109/WSC.2017.8247941
DO - 10.1109/WSC.2017.8247941
M3 - Conference contribution
SP - 2080
EP - 2091
BT - 2017 Winter Simulation Conference, WSC 2017
PB - Institute of Electrical and Electronics Engineers Inc.
ER -