SIAM Workshop on Combinatorial Scientific Computing 2020

FEBRUARY 11–13, 2020, SEATTLE, WASHINGTON, U.S.

Program

Based on the evaluation of the program committee, a subset of the following talks was accepted for publication as full papers in the proceedings.

Tuesday, February 11, 2020
8:30 - 8:45 Opening Remarks. H. M. Bücker and X. S. Li.
Invited Presentation 1 (chaired by A. Pothen)
8:45 - 9:30 Invited Presentation: Unearthing the Relationship Between Graph Isomorphism, Graph Neural Networks, and Matrix Factorization. Bruno Ribeiro, Purdue University, USA.
Session 1: Matching and Covering (chaired by J. Langguth)
9:40 - 10:05 WeGotYouCovered: The Winning Solver from the PACE 2019 Challenge, Vertex Cover Track. D. Hespe, S. Lamm, C. Schulz, and D. Strash.
10:05 - 10:30 A Parallel 2/3-Approximation Algorithm for Vertex-Weighted Matching. A. Al-Herz and A. Pothen.
10:30 - 11:00 Break
Session 2: Automatic Differentiation (chaired by U. Naumann)
11:00 - 11:25 Checkpointing Approaches Covering Resilience Issues. A. Bockhorn, S. H. K. Narayanan, and A. Walther.
11:25 - 11:50 An Inexact Combinatorial Model for Maximizing the Number of Discovered Nonzero Entries. M. A. Rostami and H. M. Bücker.
11:50 - 12:15 Algorithmic Differentiation for Machine Learning: Advanced Tracing and Handling Nonsmoothness. A. Walther, T. Kreimeier, K. Kulshreshtha, and A. Gebremedhin.
12:15 - 13:30 Break
Session 3:  Graph Algorithms, Part I (chaired by U. Catalyürek)
13:30 - 13:55 Fast Implementation of the Traveling-Salesman-Problem Method Used to Reorder Columns within Supernodes. M. Jacquelin, E. Ng, and B. Peyton.
13:55 - 14:20 A Parallel Projection Method for Metric Constrained Optimization. C. Ruggles, N. Veldt, and D. Gleich.
14:20 - 14:45 Massively Parallel Symbolic LU Factorization on GPUs. A. Gaihre, X. S. Li, and H. Liu.
14:45 - 15:30 Break
Session 4:  Graph Generation (chaired by A. J. Yzelman)
15:30 - 15:55 Parallel Generation of Uniformly-Random Simple Graphs. J. Garbus, C. Brissette, and G. M. Slota.
15:55 - 16:20 Generating Graphs to Scale for Parallel Coloring. X. Cheng, H. Maji, and A. Pothen.
Wednesday, February 12, 2020
Session 5:  Combinatorial Optimization (chaired by P. Hovland)
9:15 - 9:40 Solution Attractor of Local Search System: A Method to Reduce Computational Complexity for the Traveling Salesman Problem. W. Li.
9:40 - 10:05 A Study of Performance of Optimal Transport. Y. Dong, Y. Gao, R. Peng, I. Razenshteyn, and S. Sawlani.
10:05 - 10:30 Street-level Travel-time Estimation via Aggregated Uber Data. K. Maass, A. Sathanur, A. Khan, and R. Rallo.
10:30 - 11:00  Break
Session 6:  Partitioning, Part I (chaired by T. Davis)
11:00 - 11:25 A Multilevel Mesh Partitioning Algorithm Driven by Memory Constraints. S. Morais, C. Chevalier, and F. Ledoux.
11:25 - 11:50 Generalized Gains for Hybrid Vertex Separator Algorithms. S. Kolodziej and T. Davis.
11:50 - 12:15 Discussion, Outlook and Paper Award
12:15 - 13:30 Break
Session 7:  Partitioning, Part II (chaired by C. Aykanat)
13:30 - 13:55 Load-Balanced Bottleneck Objectives in Process Mapping. J. Langguth, S. Schlag, and C. Schulz.
13:55 - 14:20 2D Block Cyclic Partitioning for Sparse Matrices. S. Acer, E. Boman, and C. Aykanat.
14:20 - 14:45 Heuristics for Symmetric Rectilinear Matrix Partitioning. A. Yaşar and Ü. V. Çatalyürek.
14:45 - 15:30 Break
Session 8:  Linear Algebra Algorithms (chaired by S. Toledo)
15:30 - 15:55 Fault Tolerance with High Performance for Fast Matrix Multiplication. N. Birnbaum, R. Nissim, and O. Schwartz.
15:55 - 16:20 On Sparse Matrix Chain Products. U. Naumann.
16:20 - 16:45 Sparsifying the Operators of Fast Matrix Multiplication Algorithms. G. Beniamini, N. Cheng, O. Holtz, E. Karstadt, and O. Schwartz.
16:45 - 17:15 Break
Invited Presentation 2 (chaired by E. Boman)
17:15 - 18:00 Invited Presentation: Parallel Tomographic Reconstruction – Where Combinatorics Meets Geometry. Rob Bisseling, Utrecht University, The Netherlands.
18:00 - 20:00 Reception
Thursday, February 13, 2020
Invited Presentation 3 (chaired by B. Ucar)
8:30 - 9:15 Invited Presentation: Models for Optimizing Multiple Communication Cost Metrics in Parallelizing Irregular Applications. Cevdet Aykanat, Bilkent University, Turkey.
Session 9:  Direct Methods (chaired by S. Rajamanickam)
9:20 - 9:45 Matrix Symmetrization and Sparse Direct Solvers. R. Portase and B. Ucar.
9:45 - 10:10 Asymptotic Complexity of Low-rank Sparse Direct Solvers with Sparse Right-hand Sides. P. Amestoy, A. Buttari, J.-Y. L'Excellent, T. Mary, and G. Moreau.
10:10 - 10:35 Parallel Multifrontal Sparse LU Factorization Based on UMFPACK. M. M. Aznaveh and T. Davis.
10:35 - 11:00 Break
Session 10:  Numerical Methods (chaired by R. Peng)
11:00 - 11:25 Geometric Sparsification of Closeness Relations: Eigenvalue Clustering for Computing Matrix Functions. N. Goren, D. Halperin, and S. Toledo.
11:25 - 11:50 Numerical Difficulties of Combinatorial Preconditioning. K. Deweese, R. Peng, S. Stan, and H. Xu.
11:50 - 12:15 On Modeling Local Search with Special-Purpose Combinatorial Optimization Hardware. X. Liu, H. Ushijima-Mwesigwa, A. Mandal, S. Upadhyay, I. Safro, and A. Roy.
12:15 - 13:30 Break
Session 11:  Graph Algorithms, Part II (chaired by H. Meyerhenke)
13:30 - 13:55 Parallel GraphBLAS with OpenMP. M. Aznaveh, J. Chen, T. Davis, B. Hegyi, S. P. Kolodziej, T. G. Mattson, and G. Szarnyas.
13:55 - 14:20 Parallel Community Detection Using Direction-optimizing Label Propagation. X. Liu, M. Halappanavar, K. Barker, A. Lumsdaine, and A. Gebremedhin.
14:20 - 14:45 Distributed Biconnectivity. I. Bogle, G. Slota, S. Rajamanickam, and K. Devine.
14:45 - 15:30 Break
Session 12:  Network Algorithms and Appications (chaired by O. Schwarz)
15:30 - 15:55 Network Partitioning and Avoidable Contention. Y. Oltchik and O. Schwartz.
15:55 - 16:20 Modeling Fluid Mixing in Microfluidic Grids. H. Luu and M. Chrobak.
16:20 - 16:45 Approximating the Diagonal of a Graph Laplacian's Pseudoinverse, with an Application in Complex Network Analysis. E. Angriman, M. Predari, A. van der Grinten, and H. Meyerhenke.
16:45 - 18:00 Break
Session 13:  Posters (together with SIAM PP20)
18:00 - 20:00 BELLA: Berkeley Efficient Long-Read to Long-Read Aligner and Overlapper. G. Guidi, M. Ellis, D. Rokhsar, K. Yelick and A. Buluç.
Fast Approximation of Vertex Centralities in Large Real-World Networks. E. Angriman, A. van der Grinten, and H. Meyerhenke.
Doubly Stochastic Scaling, Matchings, and Permanents. F. Dufossé, K. Kaya, I. Panagiotas, and B. Ucar.
EXPLoring Algorithms INteractively (EXPLAIN): Interactive Educational Modules for Combinatorial Scientific Computing. H. M. Bücker and M. A. Rostami.
 Assignment Using Lagrangian Relaxation and Application in Ordering Sparse Matrices. S. M. Ferdous and A. Pothen.