Regularized Submodular Maximization at Scale

In this paper, we propose scalable methods for maximizing a regularized submodular function $f = g - \ell$ expressed as the difference between a monotone submodular function $g$ and a modular function $\ell$. Indeed, submodularity is inherently related to the notions of diversity, coverage, and representativeness. In particular, finding the mode of many popular probabilistic models of diversity, such as determinantal point processes, submodular probabilistic models, and strongly log-concave distributions, involves maximization of (regularized) submodular functions. Since a regularized function $f$ can potentially take on negative values, the classic theory of submodular maximization, which heavily relies on the non-negativity assumption of submodular functions, may not be applicable. To circumvent this challenge, we develop the first one-pass streaming algorithm for maximizing a regularized submodular function subject to a $k$-cardinality constraint. It returns a solution $S$ with the guarantee that $f(S)\geq(\phi^{-2}-\epsilon) \cdot g(OPT)-\ell (OPT)$, where $\phi$ is the golden ratio. Furthermore, we develop the first distributed algorithm that returns a solution $S$ with the guarantee that $\mathbb{E}[f(S)] \geq (1-\epsilon) [(1-e^{-1}) \cdot g(OPT)-\ell(OPT)]$ in $O(1/ \epsilon)$ rounds of MapReduce computation, without keeping multiple copies of the entire dataset in each round (as it is usually done). We should highlight that our result, even for the unregularized case where the modular term $\ell$ is zero, improves the memory and communication complexity of the existing work by a factor of $O(1/ \epsilon)$ while arguably provides a simpler distributed algorithm and a unifying analysis. We also empirically study the performance of our scalable methods on a set of real-life applications, including finding the mode of distributions, data summarization, and product recommendation.

[1]  Andreas Krause,et al.  From MAP to Marginals: Variational Inference in Bayesian Submodular Models , 2014, NIPS.

[2]  Silvio Lattanzi,et al.  Submodular Streaming in All its Glory: Tight Approximation, Minimum Memory and Low Adaptive Complexity , 2019, ICML.

[3]  Jeff A. Bilmes,et al.  Submodularity for Data Selection in Statistical Machine Translation , 2014 .

[4]  Ola Svensson,et al.  The one-way communication complexity of submodular maximization with applications to streaming and robustness , 2020, STOC.

[5]  T. Liggett,et al.  Negative dependence and the geometry of polynomials , 2007, 0707.2340.

[6]  Jan Vondrák,et al.  Optimal approximation for submodular and supermodular optimization with bounded curvature , 2013, SODA.

[7]  Morteza Zadimoghaddam,et al.  Randomized Composable Core-sets for Distributed Submodular Maximization , 2015, STOC.

[8]  Niv Buchbinder,et al.  Submodular Functions Maximization Problems , 2018, Handbook of Approximation Algorithms and Metaheuristics.

[9]  Laurence A. Wolsey,et al.  Best Algorithms for Approximating the Maximum of a Submodular Set Function , 1978, Math. Oper. Res..

[10]  Jan Vondrák,et al.  Fast algorithms for maximizing submodular functions , 2014, SODA.

[11]  Vahab S. Mirrokni,et al.  Maximizing Nonmonotone Submodular Functions under Matroid or Knapsack Constraints , 2009, SIAM J. Discret. Math..

[12]  Amit Chakrabarti,et al.  Submodular maximization meets streaming: matchings, matroids, and more , 2013, Math. Program..

[13]  Amin Karbasi,et al.  Do Less, Get More: Streaming Submodular Maximization with Subsampling , 2018, NeurIPS.

[14]  Jure Leskovec,et al.  {SNAP Datasets}: {Stanford} Large Network Dataset Collection , 2014 .

[15]  Amin Karbasi,et al.  Streaming Submodular Maximization under a k-Set System Constraint , 2020, ICML.

[16]  Suvrit Sra,et al.  Flexible Modeling of Diversity with Strongly Log-Concave Distributions , 2019, NeurIPS.

[17]  Roy Schwartz,et al.  A Tight Approximation for Submodular Maximization with Mixed Packing and Covering Constraints , 2018, ICALP.

[18]  Amin Karbasi,et al.  Submodular Maximization Beyond Non-negativity: Guarantees, Fast Algorithms, and Applications , 2019, ICML.

[19]  Rishabh K. Iyer,et al.  Submodularity in Data Subset Selection and Active Learning , 2015, ICML.

[20]  Eric Balkanski,et al.  The FAST Algorithm for Submodular Maximization , 2019, ICML.

[21]  Moran Feldman,et al.  Guess Free Maximization of Submodular and Linear Sums , 2018, Algorithmica.

[22]  Sergei Vassilvitskii,et al.  Fast greedy algorithms in mapreduce and streaming , 2013, SPAA.

[23]  Vahab S. Mirrokni,et al.  Maximizing Non-Monotone Submodular Functions , 2011, 48th Annual IEEE Symposium on Foundations of Computer Science (FOCS'07).

[24]  Abhimanyu Das,et al.  Submodular meets Spectral: Greedy Algorithms for Subset Selection, Sparse Approximation and Dictionary Selection , 2011, ICML.

[25]  E. Terzi,et al.  Team Formation: Striking a Balance between Coverage and Cost , 2020, 2002.07782.

[26]  Andreas Krause,et al.  Lazier Than Lazy Greedy , 2014, AAAI.

[27]  Arnaldo de Albuquerque Araújo,et al.  VSUMM: A mechanism designed to produce static video summaries and a novel evaluation method , 2011, Pattern Recognit. Lett..

[28]  Amin Karbasi,et al.  Greed Is Good: Near-Optimal Submodular Maximization via Greedy Optimization , 2017, COLT.

[29]  Leonid Gurvits A Polynomial-Time Algorithm to Approximate the Mixed Volume within a Simply Exponential Factor , 2009, Discret. Comput. Geom..

[30]  Alkis Gotovos,et al.  Sampling from Probabilistic Submodular Models , 2015, NIPS.

[31]  Rishabh K. Iyer,et al.  Submodular Point Processes with Applications to Machine learning , 2015, AISTATS.

[32]  Andreas Krause,et al.  Distributed Submodular Maximization: Identifying Representative Elements in Massive Data , 2013, NIPS.

[33]  Joseph Naor,et al.  Improved Approximations for k-Exchange Systems - (Extended Abstract) , 2011, ESA.

[34]  Huy L. Nguyen,et al.  The Power of Randomization: Distributed Submodular Maximization on Massive Datasets , 2015, ICML.

[35]  Roy Schwartz,et al.  Online Submodular Maximization with Preemption , 2015, SODA.

[36]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[37]  Baharan Mirzasoleiman,et al.  Fast Constrained Submodular Maximization: Personalized Data Summarization , 2016, ICML.

[38]  M. L. Fisher,et al.  An analysis of approximations for maximizing submodular set functions—I , 1978, Math. Program..

[39]  Jan Vondrák,et al.  Submodular Optimization in the MapReduce Model , 2018, SOSA.

[40]  F. Maxwell Harper,et al.  The MovieLens Datasets: History and Context , 2016, TIIS.

[41]  Huy L. Nguyen,et al.  An Optimal Streaming Algorithm for Non-monotone Submodular Maximization , 2019, ArXiv.

[42]  A. M. FRIEZE A cost function property for plant location problems , 1974, Math. Program..

[43]  Ola Svensson,et al.  Beyond 1/2-Approximation for Submodular Maximization on Massive Data Streams , 2018, ICML.

[44]  William Stafford Noble,et al.  Choosing panels of genomics assays using submodular optimization , 2016, Genome Biology.

[45]  Morteza Zadimoghaddam,et al.  Scalable Deletion-Robust Submodular Maximization: Data Summarization with Privacy and Fairness Constraints , 2018, ICML.

[46]  Alexandros G. Dimakis,et al.  Restricted Strong Convexity Implies Weak Submodularity , 2016, The Annals of Statistics.

[47]  Luc Van Gool,et al.  Video summarization by learning submodular mixtures of objectives , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[48]  Ben Taskar,et al.  Determinantal Point Processes for Machine Learning , 2012, Found. Trends Mach. Learn..

[49]  Rishabh K. Iyer,et al.  Submodular Optimization with Submodular Cover and Submodular Knapsack Constraints , 2013, NIPS.

[50]  Hui Lin,et al.  A Class of Submodular Functions for Document Summarization , 2011, ACL.

[51]  Amin Karbasi,et al.  A Submodular Approach to Create Individualized Parcellations of the Human Brain , 2017, MICCAI.

[52]  Morteza Zadimoghaddam,et al.  Data Summarization at Scale: A Two-Stage Submodular Approach , 2018, ICML.

[53]  Andreas Krause,et al.  Distributed Submodular Maximization , 2014, J. Mach. Learn. Res..

[54]  Amin Karbasi,et al.  Fast Mixing for Discrete Point Processes , 2015, COLT.

[55]  Rishabh K. Iyer,et al.  Learning Mixtures of Submodular Functions for Image Collection Summarization , 2014, NIPS.

[56]  Alexandros G. Dimakis,et al.  Streaming Weak Submodularity: Interpreting Neural Networks on the Fly , 2017, NIPS.

[57]  Moran Feldman,et al.  Making a Sieve Random: Improved Semi-Streaming Algorithm for Submodular Maximization under a Cardinality Constraint , 2019, ArXiv.

[58]  Andreas Krause,et al.  Streaming submodular maximization: massive data summarization on the fly , 2014, KDD.

[59]  Alkis Gotovos,et al.  Strong Log-Concavity Does Not Imply Log-Submodularity , 2019, ArXiv.

[60]  Huy L. Nguyen,et al.  A New Framework for Distributed Submodular Maximization , 2015, 2016 IEEE 57th Annual Symposium on Foundations of Computer Science (FOCS).

[61]  Kent Quanrud,et al.  Streaming Algorithms for Submodular Function Maximization , 2015, ICALP.

[62]  Moran Feldman,et al.  Optimal Streaming Algorithms for Submodular Maximization with Cardinality Constraints , 2020, ICALP.

[63]  Andreas Krause,et al.  Submodular Function Maximization , 2014, Tractability.

[64]  Alexandros G. Dimakis,et al.  Sparse and Greedy: Sparsifying Submodular Facility Location Problems , 2015 .

[65]  László Lovász,et al.  Submodular functions and convexity , 1982, ISMP.