An official website of the United States government
Reduced bases have been introduced for the approximation of parametrized PDEs in applications where many online queries are required. Their numerical efficiency for such problems has been theoretically confirmed in Binev et al. ( SIAM J. Math. Anal. 43 (2011) 1457–1472) and DeVore et al. ( Constructive Approximation 37 (2013) 455–466), where it is shown that the reduced basis space V n of dimension n , constructed by a certain greedy strategy, has approximation error similar to that of the optimal space associated to the Kolmogorov n -width of the solution manifold. The greedy construction of the reduced basis space is performed in an offline stage which requires at each step a maximization of the current error over the parameter space. For the purpose of numerical computation, this maximization is performed over a finite training set obtained through a discretization of the parameter domain. To guarantee a final approximation error ε for the space generated by the greedy algorithm requires in principle that the snapshots associated to this training set constitute an approximation net for the solution manifold with accuracy of order ε . Hence, the size of the training set is the ε covering number for M and this covering number typically behaves like exp( Cε −1/s ) for some C > 0 when the solution manifold has n -width decay O ( n −s ). Thus, the shear size of the training set prohibits implementation of the algorithm when ε is small. The main result of this paper shows that, if one is willing to accept results which hold with high probability, rather than with certainty, then for a large class of relevant problems one may replace the fine discretization by a random training set of size polynomial in ε −1 . Our proof of this fact is established by using inverse inequalities for polynomials in high dimensions.
more »
« less
Adaptive greedy algorithms based on parameter‐domain decomposition and reconstruction for the reduced basis method
Abstract The reduced basis method (RBM) empowers repeated and rapid evaluation of parametrized partial differential equations through an offline–online decomposition, a.k.a. a learning‐execution process. A key feature of the method is a greedy algorithm repeatedly scanning the training set, a fine discretization of the parameter domain, to identify the next dimension of the parameter‐induced solution manifold along which we expand the surrogate solution space. Although successfully applied to problems with fairly high parametric dimensions, the challenge is that this scanning cost dominates the offline cost due to it being proportional to the cardinality of the training set which is exponential with respect to the parameter dimension. In this work, we review three recent attempts in effectively delaying this curse of dimensionality, and propose two new hybrid strategies through successive refinement and multilevel maximization of the error estimate over the training set. All five offline‐enhanced methods and the original greedy algorithm are tested and compared on two types of problems: the thermal block problem and the geometrically parameterized Helmholtz problem.
more »
« less
- Award ID(s):
- 1719698
- PAR ID:
- 10455553
- Publisher / Repository:
- Wiley Blackwell (John Wiley & Sons)
- Date Published:
- Journal Name:
- International Journal for Numerical Methods in Engineering
- Volume:
- 121
- Issue:
- 23
- ISSN:
- 0029-5981
- Page Range / eLocation ID:
- p. 5426-5445
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
This paper investigates a novel engineering problem, i.e., security-constrained multi-period operation of micro energy-water nexuses. This problem is computationally challenging because of its high nonlinearity, nonconvexity, and large dimension. We propose a two-stage iterative algorithm employing a hybrid physics and data-driven contingency filtering (CF) method and convexification to solve it. The convexified master problem is solved in the first stage by considering the base case operation and binding contingencies set (BCS). The second stage updates BCS using physics-based data-driven methods, which include dynamic and filtered data sets. This method is faster than existing CF methods because it relies on offline optimization problems and contains a limited number of online optimization problems. We validate effectiveness of the proposed method using two different case studies: the IEEE 13-bus power system with the EPANET 8-node water system and the IEEE 33-bus power system with the Otsfeld 13-node water system.more » « less
-
We consider the problem of subset selection in the online setting, where data arrive incrementally. Instead of storing and running subset selection on the entire dataset, we propose an incremental subset selection framework that, at each time instant, uses the previously selected set of representatives and the new batch of data in order to update the set of representatives. We cast the problem as an integer bi- nary optimization minimizing the encoding cost of the data via representatives regularized by the number of selected items. As the proposed optimization is, in general, NP-hard and non-convex, we study a greedy approach based on un- constrained submodular optimization and also propose an efficient convex relaxation. We show that, under appropriate conditions, the solution of our proposed convex algorithm achieves the global optimal solution of the non-convex problem. Our results also address the conventional problem of subset selection in the offline setting, as a special case. By extensive experiments on the problem of video summarization, we demonstrate that our proposed online subset selection algorithms perform well on real data, capturing diverse representative events in videos, while they obtain objective function values close to the offline setting.more » « less
-
We consider the problem of subset selection in the online setting, where data arrive incrementally. Instead of storing and running subset selection on the entire dataset, we propose an incremental subset selection framework that, at each time instant, uses the previously selected set of representatives and the new batch of data in order to update the set of representatives. We cast the problem as an integer binary optimization minimizing the encoding cost of the data via representatives regularized by the number of selected items. As the proposed optimization is, in general, NP-hard and non-convex, we study a greedy approach based on unconstrained submodular optimization and also propose an efficient convex relaxation. We show that, under appropriate conditions, the solution of our proposed convex algorithm achieves the global optimal solution of the non-convex problem. Our results also address the conventional problem of subset selection in the offline setting, as a special case. By extensive experiments on the problem of video summarization, we demonstrate that our proposed online subset selection algorithms perform well on real data, capturing diverse representative events in videos, while they obtain objective function values close to the offline setting.more » « less
-
Bae, Sang Won; Park, Heejin (Ed.)We consider the problem of solving the Min-Sum Submodular Cover problem using local search. The Min-Sum Submodular Cover problem generalizes the NP-complete Min-Sum Set Cover problem, replacing the input set cover instance with a monotone submodular set function. A simple greedy algorithm achieves an approximation factor of 4, which is tight unless P=NP [Streeter and Golovin, NeurIPS, 2008]. We complement the greedy algorithm with analysis of a local search algorithm. Building on work of Munagala et al. [ICDT, 2005], we show that, using simple initialization, a straightforward local search algorithm achieves a (4+ε)-approximate solution in time O(n³log(n/ε)), provided that the monotone submodular set function is also second-order supermodular. Second-order supermodularity has been shown to hold for a number of submodular functions of practical interest, including functions associated with set cover, matching, and facility location. We present experiments on two special cases of Min-Sum Submodular Cover and find that the local search algorithm can outperform the greedy algorithm on small data sets.more » « less
