A data-driven robust optimization algorithm for black-box cases: An application to hyper-parameter optimization of machine learning algorithms
[Display omitted] •A novel Black-Box data-driven robust optimization approach is proposed.•A Gaussian process is used in a Bayesian optimization framework to design the approach.•The approach is consistent with the data in a predefined confidence level.•A hyper-parameter optimization for deep learni...
Saved in:
| Published in | Computers & industrial engineering Vol. 160; p. 107581 |
|---|---|
| Main Authors | , , |
| Format | Journal Article |
| Language | English |
| Published |
Elsevier Ltd
01.10.2021
|
| Subjects | |
| Online Access | Get full text |
| ISSN | 0360-8352 1879-0550 |
| DOI | 10.1016/j.cie.2021.107581 |
Cover
| Abstract | [Display omitted]
•A novel Black-Box data-driven robust optimization approach is proposed.•A Gaussian process is used in a Bayesian optimization framework to design the approach.•The approach is consistent with the data in a predefined confidence level.•A hyper-parameter optimization for deep learning is investigated as an application.•The optimal hyper-parameters are robust with respect to noise.
The huge availability of data in the last decade has raised the opportunity for the better use of data in decision-making processes. The idea of using the existing data to achieve a more coherent reality solution has led to a branch of optimization called data-driven optimization. On the one hand, the presence of uncertain variables in these datasets makes it crucial to design robust optimization methods in this area. On the other hand, in many real-world problems, the closed-form of the objective function is not available and a meta-model based framework is necessary. Motivated by the above points, in this paper a Gaussian process is used in a Bayesian optimization framework to design a method that is consistent with the data in a predefined confidence level. The advantage of the proposed method is that it is computationally tractable in addition to being robust and independent of the objective function’s form. As one of the applications of the proposed algorithm, hyper-parameter optimization for deep learning is investigated. The proposed method can help find the optimal hyper-parameters that are robust with respect to noise. |
|---|---|
| AbstractList | [Display omitted]
•A novel Black-Box data-driven robust optimization approach is proposed.•A Gaussian process is used in a Bayesian optimization framework to design the approach.•The approach is consistent with the data in a predefined confidence level.•A hyper-parameter optimization for deep learning is investigated as an application.•The optimal hyper-parameters are robust with respect to noise.
The huge availability of data in the last decade has raised the opportunity for the better use of data in decision-making processes. The idea of using the existing data to achieve a more coherent reality solution has led to a branch of optimization called data-driven optimization. On the one hand, the presence of uncertain variables in these datasets makes it crucial to design robust optimization methods in this area. On the other hand, in many real-world problems, the closed-form of the objective function is not available and a meta-model based framework is necessary. Motivated by the above points, in this paper a Gaussian process is used in a Bayesian optimization framework to design a method that is consistent with the data in a predefined confidence level. The advantage of the proposed method is that it is computationally tractable in addition to being robust and independent of the objective function’s form. As one of the applications of the proposed algorithm, hyper-parameter optimization for deep learning is investigated. The proposed method can help find the optimal hyper-parameters that are robust with respect to noise. |
| ArticleNumber | 107581 |
| Author | Azizi, Mohammad Javad Seifi, Farshad Akhavan Niaki, Seyed Taghi |
| Author_xml | – sequence: 1 givenname: Farshad surname: Seifi fullname: Seifi, Farshad email: farshad.seifi@ie.sharif.edu organization: Department of Industrial Engineering, Sharif University of Technology, Iran – sequence: 2 givenname: Mohammad Javad surname: Azizi fullname: Azizi, Mohammad Javad email: Azizim@USC.edu organization: Department of Industrial & Systems Engineering, University of Southern California, Los Angeles, CA 90007, United States – sequence: 3 givenname: Seyed Taghi surname: Akhavan Niaki fullname: Akhavan Niaki, Seyed Taghi email: Niaki@Sharif.edu organization: Department of Industrial Engineering, Sharif University of Technology, P.O. Box 11155-9414, Azadi Ave, Tehran 1458889694, Iran |
| BookMark | eNp9kE1OwzAQhS1UJNrCAdj5Ail2nF9YVRV_UiU23UcTe9y6JHFkm4pyCM5MSpEQLLoazeh9b2behIw62yEh15zNOOPZzXYmDc5iFvOhz9OCn5ExL_IyYmnKRmTMRMaiQqTxBZl4v2WMJWnJx-RzThUEiJQzO-yos_WbD9T2wbTmA4KxHYVmbZ0Jm5Zq62jdgHyNavtOJXj0t3Q-KPq-MfKoDpZu9j26qAcHLQZ0f92spi3IjemQNgiuM936d4O_JOcaGo9XP3VKVg_3q8VTtHx5fF7Ml5GMyzxEIBKBieKxUrFkulaljrVGHUulSqaFVEwVWS4F45DVkGBRJFiXiSxFkQoQU5IfbaWz3jvUlTTh-77gwDQVZ9Uh1Wo7zLE6pFodUx1I_o_snWnB7U8yd0cGh492Bl3lB0knURmHMlTKmhP0F7MWlqA |
| CitedBy_id | crossref_primary_10_1016_j_cie_2022_108028 crossref_primary_10_3934_mbe_2024275 crossref_primary_10_1016_j_iswa_2024_200390 crossref_primary_10_3390_ma17194791 crossref_primary_10_1038_s43016_022_00617_5 crossref_primary_10_1016_j_autcon_2022_104666 crossref_primary_10_1016_j_cie_2024_110492 |
| Cites_doi | 10.1287/opre.2014.1314 10.4018/978-1-930708-31-0.ch001 10.1287/opre.21.5.1154 10.1007/s10107-011-0494-7 10.1137/130925013 10.1287/mnsc.1120.1641 10.1287/educ.2015.0134 10.1080/17442508708833436 10.1038/nature14236 10.1287/opre.1030.0065 10.1137/S0895479896298130 10.1038/nature16961 10.1287/opre.1090.0741 10.1016/j.cor.2017.07.002 10.1137/050622328 10.1007/s10107-017-1125-8 10.1016/j.cor.2018.04.011 10.1287/opre.1050.0216 10.1007/s10898-012-9899-y 10.1287/moor.23.4.769 10.1515/jnma-2017-0020 10.5267/j.ijiec.2016.5.003 10.1115/1.3653121 10.1007/s10107-003-0396-4 10.1007/s10107-017-1172-1 10.1287/moor.1040.0129 |
| ContentType | Journal Article |
| Copyright | 2021 Elsevier Ltd |
| Copyright_xml | – notice: 2021 Elsevier Ltd |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.cie.2021.107581 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Applied Sciences Engineering |
| EISSN | 1879-0550 |
| ExternalDocumentID | 10_1016_j_cie_2021_107581 S036083522100485X |
| GroupedDBID | --K --M -~X .DC .~1 0R~ 1B1 1RT 1~. 1~5 29F 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JN 9JO AAAKG AABNK AACTN AAEDT AAEDW AAFWJ AAIAV AAIKC AAIKJ AAKOC AALRI AAMNW AAOAW AAQFI AAQXK AARIN AAXUO ABAOU ABMAC ABUCO ABXDB ABYKQ ACAZW ACDAQ ACGFO ACGFS ACNCT ACNNM ACRLP ADBBV ADEZE ADGUI ADMUD ADRHT ADTZH AEBSH AECPX AEKER AENEX AFKWA AFTJW AGHFR AGUBO AGYEJ AHHHB AHJVU AIEXJ AIGVJ AIKHN AITUG AJBFU AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ APLSM ARUGR ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BKOMP BLXMC CS3 DU5 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 FDB FEDTE FGOYB FIRID FNPLU FYGXN G-2 G-Q G8K GBLVA HAMUX HLZ HVGLF HZ~ H~9 IHE J1W JJJVA KOM LX9 LY1 LY7 M41 MHUIS MO0 MS~ N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. PQQKQ Q38 R2- RIG RNS ROL RPZ RXW SBC SDF SDG SDP SDS SES SET SEW SPC SPCBC SSB SSD SST SSW SSZ T5K TAE TN5 WUQ XPP ZMT ~G- AATTM AAXKI AAYWO AAYXX ABDPE ABJNI ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP CITATION EFKBS ~HD |
| ID | FETCH-LOGICAL-c297t-a343e4d12dd2c0fbd9f2ffef2cdd90f3cd0d867c301a6ba4e884eb94c93853a3 |
| IEDL.DBID | .~1 |
| ISSN | 0360-8352 |
| IngestDate | Thu Apr 24 22:57:14 EDT 2025 Thu Oct 09 00:32:52 EDT 2025 Fri Feb 23 02:43:04 EST 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Deep learning Robust optimization Gaussian process Hyper-parameter tuning Black-box optimization Bayesian optimization Data-driven optimization |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c297t-a343e4d12dd2c0fbd9f2ffef2cdd90f3cd0d867c301a6ba4e884eb94c93853a3 |
| ParticipantIDs | crossref_citationtrail_10_1016_j_cie_2021_107581 crossref_primary_10_1016_j_cie_2021_107581 elsevier_sciencedirect_doi_10_1016_j_cie_2021_107581 |
| PublicationCentury | 2000 |
| PublicationDate | October 2021 2021-10-00 |
| PublicationDateYYYYMMDD | 2021-10-01 |
| PublicationDate_xml | – month: 10 year: 2021 text: October 2021 |
| PublicationDecade | 2020 |
| PublicationTitle | Computers & industrial engineering |
| PublicationYear | 2021 |
| Publisher | Elsevier Ltd |
| Publisher_xml | – name: Elsevier Ltd |
| References | Taigman, Yang, Ranzato, Wolf (b0260) 2014 Bertsimas, Gupta, Kallus (b0045) 2018; 167 Bertsimas, Sim (b0040) 2004; 52 Liu, Wu, Xiao, Zhang (b0150) 2018; 26 Feurer, Klein, Eggensperger, Springenberg, Blum, Hutter (b0110) 2015 Xin, L., Goldberg, D. A. (2013). Time (in) consistency of multistage distributionally robust inventory models with moment constraints. arXiv preprint arXiv:1304.3074. Soyster (b0245) 1973; 21 Milford, Wyeth (b0180) 2012 Nemirovski, Shapiro (b0205) 2006; 17 Sutskever, I., Vinyals, O., Le, Q. V. (2014). Sequence to sequence learning with neural networks. In Advances in neural information processing systems (pp. 3104–3112). Chen, X., Lin, Q., Xu, G. (2019). Distributionally robust optimization with confidence bands for probability density functions. arXiv preprint arXiv:1901.02169. Rasmussen, Williams (b0220) 2005 Taguchi, G., (1986). Introduction to quality engineering: designing quality into products and processes. Asian Productivity Organization. Igel (b0130) 2005 Gao, R., Kleywegt, A. J. (2016). Distributionally robust stochastic optimization with Wasserstein distance. arXiv preprint arXiv:1604.02199. Marla, Vaze, Barnhart (b0160) 2018; 98 Zymler, Kuhn, Rustem (b0280) 2013; 137 Esfahani, Kuhn (b0095) 2018; 171 Esteban-Pérez, A., Morales, J. M. (2019). Data-driven distributionally robust optimization via optimal transport with order cone constraints. arXiv preprint arXiv:1903.01769. Shafieezadeh-Abadeh, S., Kuhn, D., Esfahani, P. M. (2017). Regularization via mass transportation. arXiv preprint arXiv:1710.10016. Dupačová (b0085) 1987; 20 Kushner (b0145) 1964; 86 (b0215) 1995 Falkner, S., Klein, A., and Hutter, F. (2018). BOHB: Robust and efficient hyperparameter optimization at scale. arXiv preprint arXiv:1807.01774. Ben-Tal, El Ghaoui, Nemirovski (b0020) 2009 Duchi, J., Glynn, P., Namkoong, H. (2016). Statistics of robust optimization: A generalized empirical likelihood approach. arXiv preprint arXiv:1610.03425. Ben-Tal, Hertog, Waegenaere, Melenberg, Rennen (b0025) 2013; 59 Ben-Tal, Nemirovski (b0015) 1998; 23 Silver, Huang, Maddison, Guez, Sifre, Van Den Driessche, Schrittwieser, Antonoglou, Panneershelvam, Lanctot (b0235) 2016; 529 Delage, Ye (b0075) 2010; 58 Marzat, Walter, Piet-Lahanier (b0165) 2013; 55 Bertsimas, Sethuraman (b0030) 2000 Iyengar (b0135) 2005; 30 Mehrotra, Papp (b0170) 2014; 24 Mendoza, Klein, Feurer, Springenberg, Urban, Burkart, Dippel, Lindauer, Hutter (b0175) 2019 Feurer, Hutter (b0115) 2019 Wiesemann, Kuhn, Sim (b0270) 2014; 62 Nakao, Shen, Chen (b0195) 2017; 88 Smith, K. A. (2002). Neural networks for business: an introduction. Neural Networks in Business: Techniques and Applications, IGI Global: 1-24. Lizotte (b0155) 2008 Wang, J., Clark, S. C., Liu, E., Frazier, P. I. (2016). Parallel bayesian global optimization of expensive functions. arXiv preprint arXiv:1602.05149. Scarf (b0225) 1957 Bertsimas, Sim (b0035) 2003; 98 Boyd, Vandenberghe (b0060) 2004 Blanchet, J., Kang, Y., Zhang, F., He, F., Hu, Z. (2017a). Doubly robust data-driven distributionally robust optimization. arXiv preprint arXiv:1705.07168. Krizhevsky, A., Sutskever, I., Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systemsconference (NIPS 2012) 1097-105. Nilim, El Ghaoui (b0210) 2005; 53 Mnih, Kavukcuoglu, Silver, Rusu, Veness, Bellemare, Graves, Riedmiller, Fidjeland, Ostrovski (b0185) 2015; 518 Moghaddam, Mahlooji (b0190) 2016; 7 El Ghaoui, Lebret (b0090) 1997; 18 Blanchet, J., Kang, Y., Zhang, F., Murthy, K. (2017b). Data-driven optimal transport cost selection for distributionally robust optimization. arXiv preprint arXiv:1705.07152. Namkoong, Duchi (b0200) 2016 Bayraksan, G., Love, D. K. (2015). Data-driven stochastic programming using phi-divergences. In The Operations Research Revolution (INFORMS), https://doi.org/10.1287/educ.2015.0134. Azizi, Seifi, Moghadam (b0005) 2019 Gao, R., Kleywegt, A. J. (2017). Distributionally robust stochastic optimization with dependence structure. arXiv preprint arXiv:1701.04200. Iyengar (10.1016/j.cie.2021.107581_b0135) 2005; 30 10.1016/j.cie.2021.107581_b0250 10.1016/j.cie.2021.107581_b0055 10.1016/j.cie.2021.107581_b0010 10.1016/j.cie.2021.107581_b0255 Mnih (10.1016/j.cie.2021.107581_b0185) 2015; 518 Liu (10.1016/j.cie.2021.107581_b0150) 2018; 26 Mehrotra (10.1016/j.cie.2021.107581_b0170) 2014; 24 Dupačová (10.1016/j.cie.2021.107581_b0085) 1987; 20 Nilim (10.1016/j.cie.2021.107581_b0210) 2005; 53 Ben-Tal (10.1016/j.cie.2021.107581_b0020) 2009 Moghaddam (10.1016/j.cie.2021.107581_b0190) 2016; 7 10.1016/j.cie.2021.107581_b0050 Marzat (10.1016/j.cie.2021.107581_b0165) 2013; 55 Marla (10.1016/j.cie.2021.107581_b0160) 2018; 98 Wiesemann (10.1016/j.cie.2021.107581_b0270) 2014; 62 Feurer (10.1016/j.cie.2021.107581_b0115) 2019 Nemirovski (10.1016/j.cie.2021.107581_b0205) 2006; 17 Bertsimas (10.1016/j.cie.2021.107581_b0045) 2018; 167 Taigman (10.1016/j.cie.2021.107581_b0260) 2014 Nakao (10.1016/j.cie.2021.107581_b0195) 2017; 88 Boyd (10.1016/j.cie.2021.107581_b0060) 2004 10.1016/j.cie.2021.107581_b0240 Rasmussen (10.1016/j.cie.2021.107581_b0220) 2005 10.1016/j.cie.2021.107581_b0120 Azizi (10.1016/j.cie.2021.107581_b0005) 2019 10.1016/j.cie.2021.107581_b0125 Mendoza (10.1016/j.cie.2021.107581_b0175) 2019 El Ghaoui (10.1016/j.cie.2021.107581_b0090) 1997; 18 Ben-Tal (10.1016/j.cie.2021.107581_b0015) 1998; 23 10.1016/j.cie.2021.107581_b0080 Kushner (10.1016/j.cie.2021.107581_b0145) 1964; 86 Delage (10.1016/j.cie.2021.107581_b0075) 2010; 58 Silver (10.1016/j.cie.2021.107581_b0235) 2016; 529 Ben-Tal (10.1016/j.cie.2021.107581_b0025) 2013; 59 10.1016/j.cie.2021.107581_b0275 10.1016/j.cie.2021.107581_b0230 Namkoong (10.1016/j.cie.2021.107581_b0200) 2016 Scarf (10.1016/j.cie.2021.107581_b0225) 1957 Bertsimas (10.1016/j.cie.2021.107581_b0030) 2000 10.1016/j.cie.2021.107581_b0070 Esfahani (10.1016/j.cie.2021.107581_b0095) 2018; 171 Bertsimas (10.1016/j.cie.2021.107581_b0040) 2004; 52 10.1016/j.cie.2021.107581_b0105 Feurer (10.1016/j.cie.2021.107581_b0110) 2015 Lizotte (10.1016/j.cie.2021.107581_b0155) 2008 (10.1016/j.cie.2021.107581_b0215) 1995 10.1016/j.cie.2021.107581_b0140 10.1016/j.cie.2021.107581_b0100 10.1016/j.cie.2021.107581_b0265 Zymler (10.1016/j.cie.2021.107581_b0280) 2013; 137 Soyster (10.1016/j.cie.2021.107581_b0245) 1973; 21 Bertsimas (10.1016/j.cie.2021.107581_b0035) 2003; 98 Igel (10.1016/j.cie.2021.107581_b0130) 2005 Milford (10.1016/j.cie.2021.107581_b0180) 2012 |
| References_xml | – reference: Taguchi, G., (1986). Introduction to quality engineering: designing quality into products and processes. Asian Productivity Organization. – start-page: 534 year: 2005 end-page: 546 ident: b0130 article-title: Multi-objective model selection for support vector machines publication-title: International Conference on Evolutionary Multi-Criterion Optimization – volume: 171 start-page: 115 year: 2018 end-page: 166 ident: b0095 article-title: Data-driven distributionally robust optimization using the Wasserstein metric: Performance guarantees and tractable reformulations publication-title: Mathematical Programming – volume: 98 start-page: 165 year: 2018 end-page: 184 ident: b0160 article-title: Robust optimization: Lessons learned from aircraft routing publication-title: Computers & Operations Research – year: 2008 ident: b0155 article-title: Practical Bayesian Optimization – reference: Chen, X., Lin, Q., Xu, G. (2019). Distributionally robust optimization with confidence bands for probability density functions. arXiv preprint arXiv:1901.02169. – volume: 59 start-page: 341 year: 2013 end-page: 357 ident: b0025 article-title: Robust solutions of optimization problems affected by uncertain probabilities publication-title: Management Science – reference: Sutskever, I., Vinyals, O., Le, Q. V. (2014). Sequence to sequence learning with neural networks. In Advances in neural information processing systems (pp. 3104–3112). – volume: 86 start-page: 97 year: 1964 end-page: 106 ident: b0145 article-title: A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise publication-title: Journal of Basic Engineering – volume: 26 start-page: 141 year: 2018 end-page: 150 ident: b0150 article-title: A note on distributionally robust optimization under moment uncertainty publication-title: Journal of Numerical Mathematics – reference: Xin, L., Goldberg, D. A. (2013). Time (in) consistency of multistage distributionally robust inventory models with moment constraints. arXiv preprint arXiv:1304.3074. – reference: Shafieezadeh-Abadeh, S., Kuhn, D., Esfahani, P. M. (2017). Regularization via mass transportation. arXiv preprint arXiv:1710.10016. – volume: 20 start-page: 73 year: 1987 end-page: 88 ident: b0085 article-title: The minimax approach to stochastic programming and an illustrative application publication-title: Stochastics: An International Journal of Probability and Stochastic Processes – volume: 53 start-page: 780 year: 2005 end-page: 798 ident: b0210 article-title: Robust control of Markov decision processes with uncertain transition matrices publication-title: Operations Research – start-page: 1701 year: 2014 end-page: 1708 ident: b0260 article-title: Deepface: Closing the gap to human-level performance in face verification publication-title: Proceedings of the IEEE conference on computer vision and pattern recognition – reference: Wang, J., Clark, S. C., Liu, E., Frazier, P. I. (2016). Parallel bayesian global optimization of expensive functions. arXiv preprint arXiv:1602.05149. – volume: 137 start-page: 167 year: 2013 end-page: 198 ident: b0280 article-title: Distributionally robust joint chance constraints with second-order moment information publication-title: Mathematical Programming – reference: Gao, R., Kleywegt, A. J. (2017). Distributionally robust stochastic optimization with dependence structure. arXiv preprint arXiv:1701.04200. – reference: Duchi, J., Glynn, P., Namkoong, H. (2016). Statistics of robust optimization: A generalized empirical likelihood approach. arXiv preprint arXiv:1610.03425. – volume: 55 start-page: 707 year: 2013 end-page: 727 ident: b0165 article-title: Worst-case global optimization of black-box functions through Kriging and relaxation publication-title: Journal of Global Optimization – reference: Blanchet, J., Kang, Y., Zhang, F., Murthy, K. (2017b). Data-driven optimal transport cost selection for distributionally robust optimization. arXiv preprint arXiv:1705.07152. – reference: Esteban-Pérez, A., Morales, J. M. (2019). Data-driven distributionally robust optimization via optimal transport with order cone constraints. arXiv preprint arXiv:1903.01769. – start-page: 1643 year: 2012 end-page: 1649 ident: b0180 article-title: SeqSLAM: Visual route-based navigation for sunny summer days and stormy winter nights publication-title: IEEE International Conference on Robotics and Automation – volume: 24 start-page: 1670 year: 2014 end-page: 1697 ident: b0170 article-title: A cutting surface algorithm for semi-infinite convex programming with an application to moment robust optimization publication-title: SIAM Journal on Optimization – volume: 30 start-page: 257 year: 2005 end-page: 280 ident: b0135 article-title: Robust dynamic programming publication-title: Mathematics of Operations Research – start-page: 1 year: 2019 end-page: 17 ident: b0005 article-title: A robust simulation optimization algorithm using kriging and particle swarm optimization: Application to surgery room optimization publication-title: Communications in Statistics-Simulation and Computation – volume: 58 start-page: 595 year: 2010 end-page: 612 ident: b0075 article-title: Distributionally robust optimization under moment uncertainty with application to data-driven problems publication-title: Operations Research – volume: 98 start-page: 49 year: 2003 end-page: 71 ident: b0035 article-title: Robust discrete optimization and network flows publication-title: Mathematical Programming – volume: 23 start-page: 769 year: 1998 end-page: 805 ident: b0015 article-title: Robust convex optimization publication-title: Mathematics of Operations Research – reference: Bayraksan, G., Love, D. K. (2015). Data-driven stochastic programming using phi-divergences. In The Operations Research Revolution (INFORMS), https://doi.org/10.1287/educ.2015.0134. – volume: 518 start-page: 529 year: 2015 ident: b0185 article-title: Human-level control through deep reinforcement learning publication-title: Nature – volume: 21 start-page: 1154 year: 1973 end-page: 1157 ident: b0245 article-title: Convex programming with set-inclusive constraints and applications to inexact linear programming publication-title: Operations Research – year: 1995 ident: b0215 publication-title: Stochastic Programming – volume: 88 start-page: 44 year: 2017 end-page: 57 ident: b0195 article-title: Network design in scarce data environment using moment-based distributionally robust optimization publication-title: Computers & Operations Research – year: 2019 ident: b0115 article-title: Hyperparameter optimization publication-title: Automated Machine Learning – year: 2004 ident: b0060 article-title: Convex optimization – reference: Gao, R., Kleywegt, A. J. (2016). Distributionally robust stochastic optimization with Wasserstein distance. arXiv preprint arXiv:1604.02199. – reference: Krizhevsky, A., Sutskever, I., Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. In Advances in neural information processing systemsconference (NIPS 2012) 1097-105. – reference: Smith, K. A. (2002). Neural networks for business: an introduction. Neural Networks in Business: Techniques and Applications, IGI Global: 1-24. – year: 2019 ident: b0175 article-title: Towards automatically-tuned deep neural networks publication-title: in, JMLR: Workshop and Conference Proceedings 1:1–8 (Springer) – start-page: 2208 year: 2016 end-page: 2216 ident: b0200 article-title: Stochastic gradient methods for distributionally robust optimization with f-divergences publication-title: Advances in Neural Information Processing Systems – volume: 529 start-page: 484 year: 2016 ident: b0235 article-title: Mastering the game of Go with deep neural networks and tree search publication-title: Nature – volume: 18 start-page: 1035 year: 1997 end-page: 1064 ident: b0090 article-title: Robust solutions to least-squares problems with uncertain data publication-title: SIAM Journal on Matrix Analysis and Applications – volume: 7 start-page: 517 year: 2016 end-page: 534 ident: b0190 article-title: Robust simulation optimization using φ-divergence publication-title: International Journal of Industrial Engineering Computations – volume: 62 start-page: 1358 year: 2014 end-page: 1376 ident: b0270 article-title: Distributionally robust convex optimization publication-title: Operations Research – year: 2015 ident: b0110 article-title: Efficient and robust automated machine learning publication-title: Advances in Neural Information Processing Systems – volume: 17 start-page: 969 year: 2006 end-page: 996 ident: b0205 article-title: Convex approximations of chance constrained programs publication-title: SIAM Journal on Optimization – year: 2009 ident: b0020 article-title: Robust optimization – year: 2005 ident: b0220 article-title: Gaussian processes for machine learning (adaptive computation and machine learning) – reference: Blanchet, J., Kang, Y., Zhang, F., He, F., Hu, Z. (2017a). Doubly robust data-driven distributionally robust optimization. arXiv preprint arXiv:1705.07168. – reference: Falkner, S., Klein, A., and Hutter, F. (2018). BOHB: Robust and efficient hyperparameter optimization at scale. arXiv preprint arXiv:1807.01774. – year: 2000 ident: b0030 article-title: Moment problems and semidefinite optimization publication-title: Handbook of Semidefinite Programming – volume: 52 start-page: 35 year: 2004 end-page: 53 ident: b0040 article-title: The price of robustness publication-title: Operations Research – volume: 167 start-page: 235 year: 2018 end-page: 292 ident: b0045 article-title: Data-driven robust optimization publication-title: Mathematical Programming – year: 1957 ident: b0225 article-title: A min-max solution of an inventory problemStudies in the mathematical theory of inventory and production – ident: 10.1016/j.cie.2021.107581_b0125 – year: 2019 ident: 10.1016/j.cie.2021.107581_b0175 article-title: Towards automatically-tuned deep neural networks publication-title: in, JMLR: Workshop and Conference Proceedings 1:1–8 (Springer) – ident: 10.1016/j.cie.2021.107581_b0070 – year: 2008 ident: 10.1016/j.cie.2021.107581_b0155 – volume: 62 start-page: 1358 year: 2014 ident: 10.1016/j.cie.2021.107581_b0270 article-title: Distributionally robust convex optimization publication-title: Operations Research doi: 10.1287/opre.2014.1314 – year: 1957 ident: 10.1016/j.cie.2021.107581_b0225 – ident: 10.1016/j.cie.2021.107581_b0230 – ident: 10.1016/j.cie.2021.107581_b0140 – ident: 10.1016/j.cie.2021.107581_b0240 doi: 10.4018/978-1-930708-31-0.ch001 – start-page: 2208 year: 2016 ident: 10.1016/j.cie.2021.107581_b0200 article-title: Stochastic gradient methods for distributionally robust optimization with f-divergences publication-title: Advances in Neural Information Processing Systems – ident: 10.1016/j.cie.2021.107581_b0050 – ident: 10.1016/j.cie.2021.107581_b0080 – year: 2005 ident: 10.1016/j.cie.2021.107581_b0220 – start-page: 534 year: 2005 ident: 10.1016/j.cie.2021.107581_b0130 article-title: Multi-objective model selection for support vector machines – volume: 21 start-page: 1154 year: 1973 ident: 10.1016/j.cie.2021.107581_b0245 article-title: Convex programming with set-inclusive constraints and applications to inexact linear programming publication-title: Operations Research doi: 10.1287/opre.21.5.1154 – volume: 137 start-page: 167 year: 2013 ident: 10.1016/j.cie.2021.107581_b0280 article-title: Distributionally robust joint chance constraints with second-order moment information publication-title: Mathematical Programming doi: 10.1007/s10107-011-0494-7 – volume: 24 start-page: 1670 year: 2014 ident: 10.1016/j.cie.2021.107581_b0170 article-title: A cutting surface algorithm for semi-infinite convex programming with an application to moment robust optimization publication-title: SIAM Journal on Optimization doi: 10.1137/130925013 – volume: 59 start-page: 341 year: 2013 ident: 10.1016/j.cie.2021.107581_b0025 article-title: Robust solutions of optimization problems affected by uncertain probabilities publication-title: Management Science doi: 10.1287/mnsc.1120.1641 – ident: 10.1016/j.cie.2021.107581_b0010 doi: 10.1287/educ.2015.0134 – ident: 10.1016/j.cie.2021.107581_b0105 – volume: 20 start-page: 73 year: 1987 ident: 10.1016/j.cie.2021.107581_b0085 article-title: The minimax approach to stochastic programming and an illustrative application publication-title: Stochastics: An International Journal of Probability and Stochastic Processes doi: 10.1080/17442508708833436 – volume: 518 start-page: 529 year: 2015 ident: 10.1016/j.cie.2021.107581_b0185 article-title: Human-level control through deep reinforcement learning publication-title: Nature doi: 10.1038/nature14236 – ident: 10.1016/j.cie.2021.107581_b0055 – year: 2019 ident: 10.1016/j.cie.2021.107581_b0115 article-title: Hyperparameter optimization – year: 2000 ident: 10.1016/j.cie.2021.107581_b0030 article-title: Moment problems and semidefinite optimization – volume: 52 start-page: 35 year: 2004 ident: 10.1016/j.cie.2021.107581_b0040 article-title: The price of robustness publication-title: Operations Research doi: 10.1287/opre.1030.0065 – year: 2009 ident: 10.1016/j.cie.2021.107581_b0020 – volume: 18 start-page: 1035 year: 1997 ident: 10.1016/j.cie.2021.107581_b0090 article-title: Robust solutions to least-squares problems with uncertain data publication-title: SIAM Journal on Matrix Analysis and Applications doi: 10.1137/S0895479896298130 – ident: 10.1016/j.cie.2021.107581_b0100 – start-page: 1701 year: 2014 ident: 10.1016/j.cie.2021.107581_b0260 article-title: Deepface: Closing the gap to human-level performance in face verification – start-page: 1 year: 2019 ident: 10.1016/j.cie.2021.107581_b0005 article-title: A robust simulation optimization algorithm using kriging and particle swarm optimization: Application to surgery room optimization publication-title: Communications in Statistics-Simulation and Computation – volume: 529 start-page: 484 year: 2016 ident: 10.1016/j.cie.2021.107581_b0235 article-title: Mastering the game of Go with deep neural networks and tree search publication-title: Nature doi: 10.1038/nature16961 – year: 2004 ident: 10.1016/j.cie.2021.107581_b0060 – year: 2015 ident: 10.1016/j.cie.2021.107581_b0110 article-title: Efficient and robust automated machine learning publication-title: Advances in Neural Information Processing Systems – ident: 10.1016/j.cie.2021.107581_b0255 – volume: 58 start-page: 595 year: 2010 ident: 10.1016/j.cie.2021.107581_b0075 article-title: Distributionally robust optimization under moment uncertainty with application to data-driven problems publication-title: Operations Research doi: 10.1287/opre.1090.0741 – volume: 88 start-page: 44 year: 2017 ident: 10.1016/j.cie.2021.107581_b0195 article-title: Network design in scarce data environment using moment-based distributionally robust optimization publication-title: Computers & Operations Research doi: 10.1016/j.cor.2017.07.002 – volume: 17 start-page: 969 year: 2006 ident: 10.1016/j.cie.2021.107581_b0205 article-title: Convex approximations of chance constrained programs publication-title: SIAM Journal on Optimization doi: 10.1137/050622328 – volume: 167 start-page: 235 year: 2018 ident: 10.1016/j.cie.2021.107581_b0045 article-title: Data-driven robust optimization publication-title: Mathematical Programming doi: 10.1007/s10107-017-1125-8 – ident: 10.1016/j.cie.2021.107581_b0275 – volume: 98 start-page: 165 year: 2018 ident: 10.1016/j.cie.2021.107581_b0160 article-title: Robust optimization: Lessons learned from aircraft routing publication-title: Computers & Operations Research doi: 10.1016/j.cor.2018.04.011 – volume: 53 start-page: 780 year: 2005 ident: 10.1016/j.cie.2021.107581_b0210 article-title: Robust control of Markov decision processes with uncertain transition matrices publication-title: Operations Research doi: 10.1287/opre.1050.0216 – start-page: 1643 year: 2012 ident: 10.1016/j.cie.2021.107581_b0180 article-title: SeqSLAM: Visual route-based navigation for sunny summer days and stormy winter nights – volume: 55 start-page: 707 year: 2013 ident: 10.1016/j.cie.2021.107581_b0165 article-title: Worst-case global optimization of black-box functions through Kriging and relaxation publication-title: Journal of Global Optimization doi: 10.1007/s10898-012-9899-y – ident: 10.1016/j.cie.2021.107581_b0250 – ident: 10.1016/j.cie.2021.107581_b0120 – volume: 23 start-page: 769 year: 1998 ident: 10.1016/j.cie.2021.107581_b0015 article-title: Robust convex optimization publication-title: Mathematics of Operations Research doi: 10.1287/moor.23.4.769 – volume: 26 start-page: 141 year: 2018 ident: 10.1016/j.cie.2021.107581_b0150 article-title: A note on distributionally robust optimization under moment uncertainty publication-title: Journal of Numerical Mathematics doi: 10.1515/jnma-2017-0020 – volume: 7 start-page: 517 year: 2016 ident: 10.1016/j.cie.2021.107581_b0190 article-title: Robust simulation optimization using φ-divergence publication-title: International Journal of Industrial Engineering Computations doi: 10.5267/j.ijiec.2016.5.003 – year: 1995 ident: 10.1016/j.cie.2021.107581_b0215 – ident: 10.1016/j.cie.2021.107581_b0265 – volume: 86 start-page: 97 year: 1964 ident: 10.1016/j.cie.2021.107581_b0145 article-title: A new method of locating the maximum point of an arbitrary multipeak curve in the presence of noise publication-title: Journal of Basic Engineering doi: 10.1115/1.3653121 – volume: 98 start-page: 49 year: 2003 ident: 10.1016/j.cie.2021.107581_b0035 article-title: Robust discrete optimization and network flows publication-title: Mathematical Programming doi: 10.1007/s10107-003-0396-4 – volume: 171 start-page: 115 year: 2018 ident: 10.1016/j.cie.2021.107581_b0095 article-title: Data-driven distributionally robust optimization using the Wasserstein metric: Performance guarantees and tractable reformulations publication-title: Mathematical Programming doi: 10.1007/s10107-017-1172-1 – volume: 30 start-page: 257 year: 2005 ident: 10.1016/j.cie.2021.107581_b0135 article-title: Robust dynamic programming publication-title: Mathematics of Operations Research doi: 10.1287/moor.1040.0129 |
| SSID | ssj0004591 |
| Score | 2.3870351 |
| Snippet | [Display omitted]
•A novel Black-Box data-driven robust optimization approach is proposed.•A Gaussian process is used in a Bayesian optimization framework to... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 107581 |
| SubjectTerms | Bayesian optimization Black-box optimization Data-driven optimization Deep learning Gaussian process Hyper-parameter tuning Robust optimization |
| Title | A data-driven robust optimization algorithm for black-box cases: An application to hyper-parameter optimization of machine learning algorithms |
| URI | https://dx.doi.org/10.1016/j.cie.2021.107581 |
| Volume | 160 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: Baden-Württemberg Complete Freedom Collection (Elsevier) customDbUrl: eissn: 1879-0550 dateEnd: 99991231 omitProxy: true ssIdentifier: ssj0004591 issn: 0360-8352 databaseCode: GBLVA dateStart: 20110101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier – providerCode: PRVESC databaseName: Elsevier ScienceDirect customDbUrl: eissn: 1879-0550 dateEnd: 99991231 omitProxy: true ssIdentifier: ssj0004591 issn: 0360-8352 databaseCode: .~1 dateStart: 19950101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier – providerCode: PRVESC databaseName: Elsevier SD Complete Freedom Collection [SCCMFC] customDbUrl: eissn: 1879-0550 dateEnd: 99991231 omitProxy: true ssIdentifier: ssj0004591 issn: 0360-8352 databaseCode: ACRLP dateStart: 19950101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier – providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals [SCFCJ] customDbUrl: eissn: 1879-0550 dateEnd: 99991231 omitProxy: true ssIdentifier: ssj0004591 issn: 0360-8352 databaseCode: AIKHN dateStart: 19950101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier – providerCode: PRVLSH databaseName: Elsevier Journals customDbUrl: mediaType: online eissn: 1879-0550 dateEnd: 99991231 omitProxy: true ssIdentifier: ssj0004591 issn: 0360-8352 databaseCode: AKRWK dateStart: 19770101 isFulltext: true providerName: Library Specific Holdings |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV07T8MwELYqWGDgUUC85YEJyTRx3DzYqoqqgNSFInWL_IQi0lRpkJj4Cfxmzo1DWwkY2GLrzon8uPscf75D6ILythIyFrDEdUxYon0Sh_DEQw14gnM77JZtMQj7j-xu1B41ULe-C2Nplc72VzZ9bq1dTcv1Zms6HrcewPZW-MEGPYvbI3uDnUU2i8HVh78UMbzKmgfCxErXJ5tzjhc0C1tE6kMZcLP_s29a8je9HbTlgCLuVN-yixp60kTbDjRityRnTbS5FFFwD312sOV8ElVYK4aLXLzNSpyDXcjchUvMX5_yYlw-ZxjwKhb2Bx4R-TuW4M9m17gDEotDbVzm-Bm2qgWxMcIzy51ZbS03OJvzMTV2CSieFm-Y7aNh72bY7ROXcoFImkQl4QELNFM-VYpKzwiVGGqMNlQqlXgmkMpTcRhJMAs8FJzpOGZaJEwmAfh9HhygtUk-0YcImwg0tA5iDVvQCHAiNZQJCRoAYBKpj5BX93UqXThymxXjNa15Zy9Qr1M7PGk1PEfo8ltlWsXi-EuY1QOYrkyoFHzF72rH_1M7QRu2VLH8TtFaWbzpM0ArpTifT8dztN65ve8PvgDZ4uyM |
| linkProvider | Elsevier |
| linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1JT-swEB6xHIAD8FjE_nzghGTaOG4WbhUCleVxeUXqLfJaimiD0iBx4ifwmxk3DhQJOHBLnBkn8tgz38TjGYBDJlpaqkTiEjcJ5akJaBLhlYgM4gkhnNhdtMVN1Lnll71WbwZO67MwLqzS6_5Kp0-0tW9p-NFsPA4Gjf-oeyv84JKeJa3eLMzzFoudB3b8EkylDK_K5iE1deT11uYkyAv7RR-RBXiPwDn42jhNGZzzVVj2SJG0q4_5AzNmtAYrHjUSvybHa7A0lVJwHV7bxAV9Ul04NUaKXD6NS5KjYhj6E5dEPPTzYlDeDQkCViLdHzwq82ei0KCNT0gbKT52tUmZkzv0VQvqkoQPXfDM595yS4aTgExDfAWK_scbxhvQPT_rnnaor7lAFUvjkoqQh4brgGnNVNNKnVpmrbFMaZ02bah0UydRrFAviEgKbpKEG5lylYZo-EW4CXOjfGS2gNgYOYwJE4M-aIxAkVnGpUIORDCpMtvQrMc6Uz4fuSuL8ZDVgWf32G4yJ56sEs82HL2zPFbJOH4i5rUAs08zKkNj8T3bzu_Y_sJCp_vvOru-uLnahUX3pAr524O5sngy-whdSnkwmZpvXFHuIQ |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+data-driven+robust+optimization+algorithm+for+black-box+cases%3A+An+application+to+hyper-parameter+optimization+of+machine+learning+algorithms&rft.jtitle=Computers+%26+industrial+engineering&rft.au=Seifi%2C+Farshad&rft.au=Azizi%2C+Mohammad+Javad&rft.au=Akhavan+Niaki%2C+Seyed+Taghi&rft.date=2021-10-01&rft.pub=Elsevier+Ltd&rft.issn=0360-8352&rft.eissn=1879-0550&rft.volume=160&rft_id=info:doi/10.1016%2Fj.cie.2021.107581&rft.externalDocID=S036083522100485X |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0360-8352&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0360-8352&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0360-8352&client=summon |