A concurrent implementation of the surrogate management framework with application to cardiovascular shape optimization
Tóm tắt
The surrogate management framework (SMF) is an effective approach for derivative-free optimization of expensive objective functions. The SMF is typically comprised of surrogate-based infill methods (SEARCH step) coupled to pattern search optimization (POLL step). Although the latter is easy to parallelize, parallelization of the SEARCH step requires surrogate-based strategies that generate multiple candidates at each iteration. The impact of such SEARCH methods on SMF performance remains poorly explored. In this paper, we extend the SMF to incorporate concurrent evaluations at the SEARCH step by comparing two different infill approaches: single search multiple error sampling and expected improvement constant liar approaches. These variants are generalized to address non-linearly constrained problems by the filter method. The proposed methods are benchmarked for different infill sizes, while accounting for the variability in initialization. We then demonstrate the proposed methods on two shape optimization problems motivated by hemodynamically-driven surgical design. Surrogate-based multiple-infill strategies outperform their single-infill counterparts for a fixed computational time budget on bound constrained problems. Insights drawn from this study have implications not only on future instances of the SMF, but also for other surrogate-based and hybrid parallel infill methods for derivative-free optimization.
Tài liệu tham khảo
Abbott WM, Megerman J, Hasson JE, L’Italien G, Warnock DF (1987) Effect of compliance mismatch on vascular graft patency. J Vasc Surg 5(2):376–382
Abraham F, Behr M, Heinkenschloss M (2005) Shape optimization in unsteady blood flow: a numerical study of non-newtonian effects. Comput Methods Biomech Biomed Eng 8:201–212
Abramson MA, Audet C (2006) Convergence of mesh adaptive direct search to second-order stationary points. SIAM J Optim 17(2):606–619
Abramson MA, Audet C, Dennis JE, Digabel SL (2009) OrthoMADS: a deterministic MADS instance with orthogonal directions. SIAM J Optim 20(2):948–966
Audet C, Dennis JE Jr (2000) A progressive barrier for derivative-free nonlinear programming. SIAM J Optim 20(1):445–472
Audet C, Dennis JE Jr (2003) Analysis of generalized pattern searches. SIAM J Optim 13(3):889–903
Audet C, Dennis JE Jr (2004a) A pattern search filter method for nonlinear programming without derivatives. SIAM J Optim 14(4):980–1010
Audet C, Dennis Jr, J.E (2004b) Mesh adaptive direct search algorithms for constrained optimization. Tech Rep G–2004–04, Les Cahiers du GERAD, École Polytechnique de Montréal, Département de Mathématiques et de Génie Industriel, C.P. 6079, Centre-ville, Montréal (Québec), H3C 3A7 Canada
Audet C, Dennis JE Jr (2006) Mesh adaptive direct search algorithms for constrained optimization. SIAM J Optim 17(1):2–11
Audet C, Dang CK, Orban D (2011) Efficient use of parallelism in algorithmic parameter optimization applications. Optim Lett 7(3):421–433
Audet C, Le Digabel S, Tribes C (2016) Dynamic scaling in the mesh adaptive direct search algorithm for blackbox optimization. Optim Eng 17(2):333–358
Bassiouny HS, White S, Glagov S, Choi E, Giddens DP, Zarins CK (1992) Anastomotic intimal hyperplasia: mechanical injury or flow induced. J Vasc Surg 15(4):708–717
Beckley MC (2015) Comparison of sampling methods for kriging models. Ph.D. thesis, University of Pretoria
Beiranvand V, Hare W, Lucet Y (2017) Best practices for comparing optimization algorithms. Optim Eng 18(4):815–848
Belitz P (2011) Applications on multi-dimensional sphere packings: derivative-free optimization. Ph.D. thesis, University of California, San Diego
Booker AJ (2000) Well-conditioned Kriging models for optimization of computer models. Mathematics and Computing Technology Report 002, Boeing Phantom Works, Seattle, WA
Booker AJ, Dennis JE Jr, Frank PD, Serafini DB, Torczon V, Trosset MW (1999) A rigorous framework for optimization of expensive functions by surrogates. Struct Optim 17(1):1–13
Bossek J (2017) Smoof: single-and multi-objective optimization test functions. R Journal 9(1):103–113
Box GE, Hunter JS, Hunter WG (2005) Statistics for experimenters: design, innovation, and discovery, vol 2. Wiley, Hoboken
Bozsak F, Gonzalez-Rodriguez D, Sternberger Z, Belitz P, Bewley T, Chomaz JM, Barakat AI (2015) Optimization of drug delivery by drug-eluting stents. PLoS One 10(6):e0130182
Breiman L, Cutler A (1993) A deterministic algorithm for global optimization. Math Program 58(1–3):179–199
Brochu E, Cora VM, de Freitas N (2010) A tutorial on bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning. http://arxiv.org/abs/1012.2599
Chiu JJ, Chien S (2011) Effects of disturbed flow on vascular endothelium: pathophysiological basis and clinical perspectives. Physiol Rev 91(1):327–387
Contal E, Buffoni D, Robicquet A, Vayatis N (2013) Parallel Gaussian process optimization with upper confidence bound and pure exploration. In: Joint European conference on machine learning and knowledge discovery in databases. Springer, Berlin, pp 225–240
Couckuyt I, Deschrijver D, Dhaene T (2013) Fast calculation of multiobjective probability of improvement and expected improvement criteria for pareto optimization. J Global Optim 60(3):575–594
Deb K, Thiele L, Laumanns M, Zitzler E (2005) Scalable test problems for evolutionary multobjective optimization. In: Abraham A, Jain L, Goldber R (eds) Evolutionary multiobjective optimization. Springer, London, pp 105–145
Diamond P, Armstrong M (1984) Robustness of variograms and conditioning of kriging matrices. J Int Assoc Math Geol 16(8):809–822
Digabel SL (2011) Algorithm 909: NOMAD: nonlinear optimization with the MADS algorithm. ACM Trans Math Softw 37(4):1–15
Dolan ED, Moré JJ (2002) Benchmarking optimization software with performance profiles. Math Program 91(2):201–213
Dyke BV, Asaki TJ (2013) Using QR decomposition to obtain a new instance of mesh adaptive direct search with uniformly distributed polling directions. J Optim Theory Appl 159(3):805–821
Esmaily-Moghadam M, Bazilevs Y, Marsden AL (2013) A new preconditioning technique for implicitly coupled multidomain simulations with applications to hemodynamics. Comput Mech 52(5):1141–1152
Esmaily-Moghadam M, Bazilevs Y, Marsden A (2015) Impact of data distribution on the parallel performance of iterative linear solvers with emphasis on cfd of incompressible flows. Comput Mech 55(1):93–103
Esmaily-Moghadam M, Bazilevs Y, Marsden AL (2015) A bi-partitioned iterative algorithm for solving linear systems arising from incompressible flow problems. Comput Methods Appl Mech Eng 286:40–62
Feinstein JA, Benson DW, Dubin AM, Cohen MS, Maxey DM, Mahle WT, Pahl E, Villafañe J, Bhatt AB, Peng LF et al (2012) Hypoplastic left heart syndrome: current considerations and expectations. J Am College Cardiol 59(1 Supplement):S1–S42
Ginsbourger D, Le Riche R, Carraro L (2010) Kriging is well-suited to parallelize optimization. In: Tenne Y, Goh C (eds) Computational intelligence in expensive optimization problems. Adaptation learning and optimization, vol 2. Springer, Berlin
Ginsbourger D, Janusevskis J, Le Riche R (2011) Dealing with asynchronicity in parallel Gaussian Process based global optimization. Tech Rep hal-00507632
Gould N, Scott J (2016) A note on performance profiles for benchmarking software. ACM Trans Math Softw (TOMS) 43(2):15
Gramacy RB, Gray GA, Le Digabel S, Lee HK, Ranjan P, Wells G, Wild SM (2016) Modeling an augmented lagrangian for blackbox constrained optimization. Technometrics 58(1):1–11
Grechy L, Iori F, Corbett R, Shurey S, Gedroyc W, Duncan N, Caro C, Vincent P (2017) Suppressing unsteady flow in arterio-venous fistulae. Phys Fluids 29(10):101901
Griffiths G, Nagy J, Black D, Stonebridge P (2004) Randomized clinical trial of distal anastomotic interposition vein cuff in infrainguinal polytetrafluoroethylene bypass grafting. Br J Surg 91(5):560–562
Gropp W, Thakur R, Lusk E (1999) Using MPI-2: advanced features of the message passing interface. MIT Press, Cambridge
Haftka RT, Villanueva D, Chaudhuri A (2016) Parallel surrogate-assisted global optimization with expensive functions–a survey. Struct Multidiscip Optim 54(1):3–13
Haimovici H, Ascer E, Hollier L, Strandness D Jr, Towne J (1996) Haimovici’s vascular surgery. Blackwell Science, Hoboken
Hansen N (2006) The CMA evolution strategy: a comparing review. In: Lozano JA, Larranaga P, Inza I, Bengoetxea E (eds) Towards a new evolutionary computation. Springer, Berlin, pp 75–102
Hough PD, Kolda TG, Torczon VJ (2001) Asynchronous parallel pattern search for nonlinear optimization. SIAM J Sci Comput 23(1):134–156
How T, Rowe C, Gilling-Smith G, Harris P (2000) Interposition vein cuff anastomosis alters wall shear stress distribution in the recipient artery. J Vasc Surg 31(5):1008–1017
Jamil M, Yang XS (2013) A literature survey of benchmark functions for global optimization problems. Int J Math Model Numer Optim 4(2):150–94
Jansen KE, Whiting CH, Hulbert GM (2000) A generalized-\(\alpha\) method for integrating the filtered navier-stokes equations with a stabilized finite element method. Comput Meth Appl Mech Eng 190(3–4):305–319
Jin OA, Chen W, Sudijianto A (2005) An efficient algorithm for constructing optimal design of computer experiments. J Stat Plan Inference 134:268–87
Jones DR (2001) A taxonomy of global optimization methods based on response surfaces. J Global Optim 21(4):345–383
Jones DR, Perttunen CD, Stuckman BE (1993) Lipschitzian optimization without the lipschitz constant. J Optim Theory Appl 79(1):157–181
Jones DR, Schonlau M, Welch WJ (1998) Efficient global optimization of expensive black-box functions. J Global Optim 13(4):455–492
Koziel S, Michalewicz Z (1999) Evolutionary algorithms, homomorphous mappings and constrained parameter optimization. J Evol Comp 7(1):19–44
Krige D (1951) A statistical approach to some mine valuations and allied problems in the Witwatersrand. J Chem Metall Min Soc South Africa 52:119–139
Ku JP, Elkins CJ, Taylor CA (2005) Comparison of CFD and MRI flow and velocities in an in vitro large artery bypass graft model. Ann Biomed Eng 33(3):257–269
Lemson M, Tordoir J, Daemen M, Kitslaar P (2000) Intimal hyperplasia in vascular grafts. Eur J Vasc Endovasc Surg 19(4):336–350
Levy AV, Montalvo A (1985) The tunneling algorithm for the global minimization of functions. SIAM J Sci Stat Comput 6(1):15–29
Li R, Sudjianto A (2005) Analysis of computer experiments using penalized likelihood in gaussian kriging models. Technometrics 47(2):111–120
Li C, Brezillon J, Görtz S (2014) Efficient global optimization of a natural laminar airfoil based on surrogate modeling. In: Dillmann A, Heller G, Krämer E, Kreplin H, Nitsche W, Rist U (eds) New results in numerical and experimental fluid mechanics IX. Springer, Berlin, pp 53–63
Liu J, Han Z, Song W (2012) Comparison of infill sampling criteria in kriging-based aerodynamic optimization. In: 28th congress of the international council of the aeronautical sciences, pp 23–28
Longest P, Kleinstreuer C, Archie JP (2003) Particle hemodynamics analysis of miller cuff arterial anastomosis. J Vasc Surg 38(6):1353–1362
Marsden AL, Wang M, Dennis Jr JE (2003) Constrained aeroacoustic shape optimization using the surrogate management framework. In: Annual research briefs. Center for Turbulence Research, Stanford University, pp 399–412
Marsden AL, Feinstein JA, Taylor CA (2008) A computational framework for derivative-free optimization of cardiovascular geometries. Comput Meth Appl Mech Eng 197(21–24):1890–1905
Marsden AL (2014) Optimization in cardiovascular modeling. Annu Rev Fluid Mech 46:519–46
Marsden AL, Esmaily-Moghadam M (2015) Multiscale modeling of cardiovascular flows for clinical decision support. Appl Mech Rev 67(030804):1–11
McKay MD, Conover WJ, Beckman RJ (1979) A comparison of three methods for selecting values of input variables in the analysis of output from a computer code. Technometrics 21:239–245
Miettinen K (1999) Nonlinear multiobjective optimization, vol 12. International series in operations research and management science. Kluwer Academic Publishers, London
Miller J, Foreman R, Ferguson L, Faris I (1984) Interposition vein cuff for anastomosis of prosthesis to small artery. Aust N Z J Surg 54(3):283–285
Moghadam ME, Marsden TYHA (2015) The assisted bidirectional Glenn: a novel surgical approach for first-stage single-ventricle heart palliation. J Thorac Cardio Surg 149(3):699–705
Moghadam ME, Bazilevs Y, Hsia TY, Vignon-Clementel I, Marsden A (2011) A comparison of outlet boundary treatments for prevention of backflow divergence with relevance to blood flow simulations. Comput Mech 48:277–291
Neville RF, Tempesta B, Sidawy AN (2001) Tibial bypass for limb salvage using polytetrafluoroethylene and a distal vein patch. J Vasc Surg 33(2):266–272
Niederreiter H (1992) Random number generation and quasi-Monte Carlo methods, vol 63. SIAM, Philadelphia
Norberto JJ, Sidawy AN, Trad KS, Jones BA, Neville RF, Najjar SF, Sidawy MK, DePalma RG (1995) The protective effect of vein cuffed anastomoses is not mechanical in origin. J Vasc Surg 21(4):558–566
Norwood W, Kirklin J, Sanders S (1981) Hypoplastic left heart syndrome: experience with palliative surgery. J Thorac Cardiovasc Surg 82:511–9
Norwood W, Lang P, Castaneda A, Campbell D (1981) Experience with operations for hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 82:511–9
Panneton JM, Hollier LH, Hofer JM (2004) Multicenter randomized prospective trial comparing a pre-cuffed polytetrafluoroethylene graft to a vein cuffed polytetrafluoroethylene graft for infragenicular arterial bypass. Ann Vasc Surg 18(2):199–206
Park C, Haftka RT, Kim NH (2017) Remarks on multi-fidelity surrogates. Struct Multidiscip Optim 55(3):1029–1050
Passerini AG, Milsted A, Rittgers SE (2003) Shear stress magnitude and directionality modulate growth factor gene expression in preconditioned vascular endothelial cells. J Vasc Surg 37(1):182–190
Peherstorfer B, Willcox K, Gunzburger M (2018) Survey of multifidelity methods in uncertainty propagation, inference, and optimization. SIAM Rev 60(3):550–591
Peng CY, Wu C (2014) On the choice of nugget in kriging modeling for deterministic computer experiments. J Comput Graph Stat 23(1):151–168
Powell MJ (2006) The newuoa software for unconstrained optimization without derivatives. In: Large-scale nonlinear optimization. Springer, Berlin, pp 255–297
Ramachandra AB, Sankaran S, Humphrey JD, Marsden AL (2015) Computational simulation of the adaptive capacity of vein grafts in response to increased pressure. J Biomech Eng 137(3):031009
Rasmussen CE, Williams CKI (2005) Gaussian processes for machine learning (Adaptive computation and machine learning). The MIT Press, Cambridge
Rozza G (2005) On optimization, control and shape design for an arterial bypass. Int J Numer Methods Fluids 47:1411–1419
Sacks J, Welch WJ, Mitchell TJ, Wynn HP (1989) Design and analysis of computer experiments. Stat Sci 4(4):409–423
Sankaran S (2009) Stochastic optimization using a sparse grid collocation scheme. Prob Eng Mech 24(3):382–396
Santner TJ, Williams BJ, Notz WI (2003) The design and analysis of computer experiments. Springer, New York
Sasena, M.J.: Flexibility and efficiency enhancements for constrained global design optimization with kriging approximations. Ph.D. thesis, University of Michigan, Ann Arbor (2002)
Saxena DK, Duro JA, Tiwari A, Deb K, Zhang Q (2012) Objective reduction in many-objective optimization: linear and nonlinear algorithms. IEEE Trans Evol Comput 17(1):77–99
Schroeder WJ, Lorensen B, Martin K (2004) The visualization toolkit: an object-oriented approach to 3D graphics. Kitware, Clifton Park
Seo J, Schiavazzi D, Marsden A (2019) Performance of preconditioned iterative linear solvers for cardiovascular simulations in rigid and deformable vessels. Comput Mech 64(3):717–739
Shang JK, Esmaily M, Verma A, Reinhartz O, Figliola RS, Hsia TY, Feinstein JA, Marsden AL (2019) Patient-specific multiscale modeling of the assisted bidirectional Glenn. Ann Thorac Surg 107(4):1232–39
Shephard MS, Georges MK (1991) Automatic three-dimensional mesh generation by the finite octree technique. Int J Numer Methods Eng 32(4):709–749
Siegman F (1979) Use of the venous cuff for graft anastomosis. Surg Gynecol Obstet 148(6):930–930
Stonebridge P, Prescott R, Ruckley C (1997) Randomized trial comparing infrainguinal polytetrafluoroethylene bypass grafting with and without vein interposition cuff at the distal anastomosis. J Vasc Surg 26(4):543–550
Tamisier D, Vouhe P, Vernant F, Leca F, Massot C, Neveux J (1990) Modified blalock-taussig shunts: results in infants less than 3 months of age. Ann Thorac Surg 49:797–801
Taylor CA, Hughes TJR, Zarins CK (1998) Finite element modeling of blood flow in arteries. Comput Meth Appl Mech Eng 158:155–196
Torczon V, Trosset MW (1998) From evolutionary operation to parallel direct search: pattern search algorithms for numerical optimization. Comput Sci Stat 29:396–401
Towns J, Cockerill T, Dahan M, Foster I, Gaither K, Grimshaw A, Hazlewood V, Lathrop S, Lifka D, Peterson GD et al (2014) XSEDE: accelerating scientific discovery. Comput Sci Eng 16(5):62–74
Updegrove A, Wilson NM, Merkow J, Lan H, Marsden A, Shadden S (2016) Simvascular: an open source pipeline for cardiovascular simulation. Ann Biomed Eng 45(3):525–541
Veith FJ, Gupta SK, Ascer E, White-Flores S, Samson RH, Scher LA, Towne JB, Bernhard VM, Bonier P, Flinn WR, Astelford P, Yao JS, Bergan JJ (1986) Six-year prospective multicenter randomized comparison of autologous saphenous vein and expanded polytetrafluoroethylene grafts in infrainguinal arterial reconstructions. J Vasc Surg 3(1):104–114
Verma A, Esmaily M, Shang J, Figliola R, Feinstein JA, Hsia TY, Marsden AL (2018) Optimization of the assisted bidirectional glenn procedure for first stage single ventricle repair. World J Pediatr Congenit Heart Surg 9(2):157–170
Vignon-Clementel IE, Figueroa CA, Jansen KE, Taylor CA (2006) Outflow boundary conditions for three-dimensional finite element modeling of blood flow and pressure in arteries. Comput Meth Appl Mech Eng 195:3776–3796
Whiting CH, Jansen KE (2001) A stabilized finite element method for the incompressible Navier–Stokes equations using a hierarchical basis. Int J Numer Meth Fluid 35(1):93–116
Wild S, More J (2009) Benchmarking derivative-free optimization algorithms. SIAM J Optim 20(1):172–191
Wilson N, Wang K, Dutton R, Taylor CA (2001) A software framework for creating patient specific geometric models from medical imaging data for simulation based medical planning of vascular surgery. Lect Notes Comput Sci 2208:449–456
Wong K, Brown L, Coan J, White D (2014) Distributive interoperable executive library (DIEL) for systems of multiphysics simulation. In: 2014 15th international conference on parallel and distributed computing, applications and technologies. IEEE
Yang W, Feinstein JA, Marsden AL (2010) Constrained optimization of an idealized Y-shaped baffle for the Fontan surgery at rest and exercise. Comput Methods Appl Mech Eng 199:2135–2149
Yang W, Vignon-Clementel IE, Troianowski G, Reddy VM, Feinstein JA, Marsden AL (2012) Hepatic blood flow distribution and performance in conventional and novel y-graft fontan geometries: a case series computational fluid dynamics study. J Thorac Cardiovasc Surg 143(5):1086–1097