Abstract
Maximizing a non-negative, monontone, submodular function \(f\) over \(n\) elements under a cardinality constraint \(k\) (SMCC) is a well-studied NP-hard problem. It has important applications in, e.g., machine learning and influence maximization. Though the theoretical problem admits polynomial-time approximation algorithms, solving it in practice often involves frequently querying submodular functions that are expensive to compute. This has motivated significant research into designing parallel approximation algorithms in the adaptive complexity model; adaptive complexity (adaptivity) measures the number of sequential rounds of \({{\,\textrm{poly}\,}}(n)\) function queries an algorithm requires. The state-of-the-art algorithms can achieve \((1-\frac{1}{e}-\varepsilon )\)-approximate solutions with \(O(\frac{1}{\varepsilon ^2}\log n)\) adaptivity, which approaches the known adaptivity lower-bounds. However, the \(O(\frac{1}{\varepsilon ^2} \log n)\) adaptivity only applies to maximizing worst-case functions that are unlikely to appear in practice. Thus, in this paper, we consider the special class of \(p\)-superseparable submodular functions, which places a reasonable constraint on \(f\), based on the parameter \(p\), and is more amenable to maximization, while also having real-world applicability. Our main contribution is the algorithm LS+GS, a finer-grained version of the existing LS+PGB algorithm, designed for instances of SMCC when \(f\) is \(p\)-superseparable; it achieves an expected \((1-\frac{1}{e}-\varepsilon )\)-approximate solution with \(O(\frac{1}{\varepsilon ^2}\log (pk))\) adaptivity independent of \(n\). Additionally, unrelated to \(p\)-superseparability, our LS+GS algorithm uses only \(O({\varepsilon }^{-1}{n} + {\varepsilon ^{-2}}\log n)\) oracle queries, which has an improved dependence on \(\varepsilon ^{-1}\) over the state-of-the-art LS+PGB; this is achieved through the design of a novel thresholding subroutine.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Our work does not focus on the other two classes of \(p\)-separable functions, which are \(p\)-subseparable and rev-\(p\)-subseparable functions.
References
Balkanski, E., Rubinstein, A., Singer, Y.: An optimal approximation for submodular maximization under a matroid constraint in the adaptive complexity model. In: 51st STOC, pp. 66–77. ACM (2019)
Balkanski, E., Rubinstein, A., Singer, Y.: An exponential speedup in parallel running time for submodular maximization without loss in approximation. In: 30th SODA, pp. 283–302. ACM-SIAM (2019)
Balkanski, E., Singer, Y.: The adaptive complexity of maximizing a submodular function. In: 50th STOC, pp. 1138–1151. ACM (2018)
Balkanski, E., Singer, Y.: Approximation guarantees for adaptive sampling. In: 35th ICML, pp. 384–393. PMLR (2018)
Breuer, A., Balkanski, E., Singer, Y.: The FAST algorithm for submodular maximization. In: 37th ICML, pp. 1134–1143. PMLR (2020)
Chekuri, C., Quanrud, K.: Submodular function maximization in parallel via the multilinear relaxation. In: 30th SODA, pp. 303–322. ACM-SIAM (2019)
Chen, Y., Dey, T., Kuhnle, A.: Best of both worlds: practical and theoretically optimal submodular maximization in parallel. In: Advances in Neural Information Processing Systems, vol. 34, pp. 25528–25539 (2021)
Dueck, D., Frey, B.J.: Non-metric affinity propagation for unsupervised image categorization. In: 11th CCV, pp. 1–8. IEEE (2007)
Ene, A., Nguyen, H.L.: Submodular maximization with nearly-optimal approximation and adaptivity in nearly-linear time. In: 30th SODA, pp. 274–282. ACM-SIAM (2019)
Fahrbach, M., Mirrokni, V., Zadimoghaddam, M.: Submodular maximization with nearly optimal approximation, adaptivity and query complexity. In: 30th SODA, pp. 255–273. ACM-SIAM (2019)
Grossman, T., Wool, A.: Computational experience with approximation algorithms for the set covering problem. Eur. J. Oper. Res. 101(1), 81–92 (1997)
Kazemi, E., Mitrovic, M., Zadimoghaddam, M., Lattanzi, S., Karbasi, A.: Submodular streaming in all its glory: tight approximation, minimum memory and low adaptive complexity. In: 36th ICML, pp. 3311–3320. PMLR (2019)
Kempe, D., Kleinberg, J., Tardos, É.: Maximizing the spread of influence through a social network. In: 9th KDD, pp. 137–146. ACM (2003)
Khanna, R., Elenberg, E., Dimakis, A., Negahban, S., Ghosh, J.: Scalable greedy feature selection via weak submodularity. In: 20th AISTATS, pp. 1560–1568. PMLR (2017)
Krause, A., Guestrin, C.: Submodularity and its applications in optimized information gathering. ACM Trans. Intell. Syst. Technol. (TIST) 2(4), 1–20 (2011)
Krause, A., Singh, A., Guestrin, C.: Near-optimal sensor placements in Gaussian processes: theory, efficient algorithms and empirical studies. J. Mach. Learn. Res. 9(2) (2008)
Kuhnle, A.: Quick streaming algorithms for maximization of monotone submodular functions in linear time. In: 24th AISTATS, pp. 1360–1368. PMLR (2021)
Leskovec, J., Krause, A., Guestrin, C., Faloutsos, C., VanBriesen, J., Glance, N.: Cost-effective outbreak detection in networks. In: 13th KDD, pp. 420–429. ACM (2007)
Li, W., Liu, P., Vondrák, J.: A polynomial lower bound on adaptive complexity of submodular maximization. In: 52nd STOC, pp. 140–152. ACM (2020)
Mirzasoleiman, B., Badanidiyuru, A., Karbasi, A.: Fast constrained submodular maximization: personalized data summarization. In: 33rd ICML, pp. 1358–1367. PMLR (2016)
Nemhauser, G.L., Wolsey, L.A.: Best algorithms for approximating the maximum of a submodular set function. Math. Oper. Res. 3(3), 177–188 (1978)
Nemhauser, G.L., Wolsey, L.A., Fisher, M.L.: An analysis of approximations for maximizing submodular set functions-I. Math. Program. 14(1), 265–294 (1978)
Skowron, P.: FPT approximation schemes for maximizing submodular functions. Inf. Comput. 257, 65–78 (2017)
Wei, K., Iyer, R., Bilmes, J.: Submodularity in data subset selection and active learning. In: 32nd ICML, pp. 1954–1963. PMLR (2015)
Acknowledgements
This work was in part supported by ARC Discovery Early Career Researcher Award (DECRA) DE190101118 and the University of Melbourne Faculty of Engineering and Information Technology, and School of Computing and Information Systems.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Cervenjak, P., Gan, J., Wirth, A. (2023). Fast Parallel Algorithms for Submodular p-Superseparable Maximization. In: Byrka, J., Wiese, A. (eds) Approximation and Online Algorithms . WAOA 2023. Lecture Notes in Computer Science, vol 14297. Springer, Cham. https://doi.org/10.1007/978-3-031-49815-2_16
Download citation
DOI: https://doi.org/10.1007/978-3-031-49815-2_16
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-49814-5
Online ISBN: 978-3-031-49815-2
eBook Packages: Computer ScienceComputer Science (R0)