Daniel Dadush
ABSTRACT
In this work, we give a novel algorithm for computing dense lattice subspaces, a conjecturally tight characterization of the lattice covering radius, and provide a bound on the slicing constant of lattice Voronoi cells. Our work is motivated by the pursuit of faster algorithms for integer programming, for which we give a conditional speedup based on the recent resolution of the ℓ2 Kannan-Lovász conjecture. Through these results, we hope to motivate further study of the interplay between the recently developed reverse Minkowski theory, lattice algorithms and convex geometry.
On the algorithmic side, our main contribution is a 2O(n)-time algorithm for computing a O(Cη(n))-approximate sublattice of minimum normalized determinant on any n-dimensional lattice, where Cη(n) = O(logn) is the reverse Minkowski constant in dimension n. Our method for finding dense lattice subspaces is surprisingly simple: we iteratively descend to a random co-dimension 1 subspace chosen to be the orthogonal space to a discrete Gaussian sample from the dual lattice. Applying this algorithm within a “filtration reduction” scheme, we further show how to compute a O(Cη(n))-approximate canonical filtration of any lattice, which corresponds to a canonical way of decomposing a lattice into dense blocks. As a primary application, we get the first 2O(n)-time algorithm for computing a sparse lattice projection whose “volume radius” provides a lower bound on the lattice covering radius that is tight within a O(log2.5 n)-factor. This provides an efficient algorithmic version of the ℓ2 Kannan-Lovász conjecture, which was recently resolved by Regev and Stephens-Davidowitz (STOC ’2017).
On the structural side, we prove a new lower bound on the covering radius which combines volumetric lower bounds across a chain of lattice projections. Assuming Bourgain’s slicing conjecture restricted to Voronoi cells of stable lattices, our lower bound implies (somewhat surprisingly) that the problem of approximating the lattice covering radius to within a constant factor is in . Complementing this result, we show that the slicing constant of any n-dimensional Voronoi cell is bounded by O(CKL,2(n)) = O(log1.5 n), the ℓ2 Kannan-Lovász constant, which complements the O(logn) bound of Regev and Stephens-Davidowitz for stable Voronoi cells.
- Divesh Aggarwal, Daniel Dadush, Oded Regev, and Noah Stephens-Davidowitz. 2015.Google Scholar
- Solving the Shortest Vector Problem in 2 n Time via Discrete Gaussian Sampling. In STOC. Available at http://arxiv.org/abs/1412.7994.Google Scholar
- Divesh Aggarwal, Daniel Dadush, and Noah Stephens-Davidowitz. 2015. Solving the Closest Vector Problem in 2 n Time–The Discrete Gaussian Strikes Again!. In IEEE 56th Annual Symposium on Foundations of Computer Science (FOCS). IEEE, 563–582. Google ScholarDigital Library
- Divesh Aggarwal and Noah Stephens-Davidowitz. 2018. Just Take the Average! An Embarrassingly Simple 2 n -Time Algorithm for SVP (and CVP). In 1st Symposium on Simplicity in Algorithms (SOSA 2018). Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik.Google Scholar
- Miklós Ajtai, Ravi Kumar, and D. Sivakumar. 2001. A Sieve Algorithm for the Shortest Lattice Vector Problem. In STOC. 601–610. 380752.380857 Google ScholarDigital Library
- Miklós Ajtai, Ravi Kumar, and D. Sivakumar. 2002. Sampling short lattice vectors and the closest lattice vector problem. In CCC. 41–45. Google ScholarDigital Library
- László Babai. 1986.Google Scholar
- On Lovász’ lattice reduction and the nearest lattice point problem. Combinatorica 6, 1 (1986), 1–13. Preliminary version in STACS 1985.Google Scholar
- W. Banaszczyk. 1993. New bounds in some transference theorems in the geometry of numbers. Math. Ann. 296, 4 (1993), 625–635. BF01445125Google ScholarCross Ref
- Daniel Dadush. 2012.Google Scholar
- Integer Programming, Lattice Algorithms, and Deterministic Volume Estimation. Ph.D. Dissertation. Georgia Institute of Technology.Google Scholar
- Daniel Dadush. 2014. A randomized sieving algorithm for approximate integer programming. Algorithmica 70, 2 (2014), 208–244. s00453- 013- 9834- 8 Preliminary version in LATIN 2012.Google ScholarDigital Library
- Daniel Dadush and Gábor Kun. 2016. Lattice sparsification and the approximate closest vector problem. Theory of Computing 12 (2016), Paper No. 2, 34. Preliminary version in SODA 2012.Google Scholar
- Daniel Dadush and Daniele Micciancio. 2013. Algorithms for the densest sublattice problem. In Proceedings of the Twenty-Fourth Annual ACM-SIAM Symposium on Discrete Algorithms. SIAM, Philadelphia, PA, 1103–1122.Google ScholarCross Ref
- Daniel Dadush, Chris Peikert, and Santosh Vempala. 2011. Enumerative lattice algorithms in any norm via M-ellipsoid coverings. In 2011 IEEE 52nd Annual Symposium on Foundations of Computer Science—FOCS 2011. IEEE Computer Soc., Los Alamitos, CA, 580–589.Google ScholarDigital Library
- Daniel Dadush and Oded Regev. 2016.Google Scholar
- Towards Strong Reverse Minkowski-Type Inequalities for Lattices. In IEEE 57th Annual Symposium on Foundations of Computer Science (FOCS).Google Scholar
- Venkatesan Guruswami, Daniele Micciancio, and Oded Regev. 2005. The complexity of the covering radius problem. computational complexity 14, 2 (Jun 2005), 90–121. Google ScholarDigital Library
- Ishay Haviv and Oded Regev. 2006. Hardness of the covering radius problem on lattices. In 21st Annual IEEE Conference on Computational Complexity (CCC’06). IEEE, 14–pp. Google ScholarDigital Library
- Robert Hildebrand and Matthias Köppe. 2013.Google Scholar
- A new Lenstra-type algorithm for quasiconvex polynomial integer minimization with complexity 2 O (n log n). Discrete Optimization 10, 1 (2013), 69 – 84. 11.003Google Scholar
- J. Håstad. 1988.Google Scholar
- Dual vectors and lower bounds for the nearest lattice point problem. Combinatorica 8, 1 (1988), 75–81.Google Scholar
- Ravi Kannan. 1987. Minkowski’s convex body theorem and integer programming. Math. Oper. Res. 12, 3 (1987), 415–440.Google ScholarCross Ref
- Ravi Kannan and László Lovász. 1988. Covering minima and lattice-point-free convex bodies. Ann. of Math. (2) 128, 3 (1988), 577–602. 1971436Google Scholar
- A. Khinchine. 1948. A quantitative formulation of the approximation theory of Kronecker. Izvestiya Akad. Nauk SSSR. Ser. Mat. 12 (1948), 113–122.Google Scholar
- B. Klartag. 2006.Google Scholar
- On convex perturbations with a bounded isotropic constant. Geom. Funct. Anal. 16, 6 (2006), 1274–1290.Google Scholar
- s00039- 006- 0588- 1Google Scholar
- J. C. Lagarias, H. W. Lenstra, Jr., and C.-P. Schnorr. 1990. Korkin-Zolotarev bases and successive minima of a lattice and its reciprocal lattice. Combinatorica 10, 4 (1990), 333–348.Google ScholarCross Ref
- A. K. Lenstra, H. W. Lenstra, Jr., and L. Lovász. 1982. Factoring polynomials with rational coefficients. Math. Ann. 261, 4 (1982), 515–534. BF01457454Google ScholarCross Ref
- H. W. Lenstra, Jr. 1983. Integer programming with a fixed number of variables. Math. Oper. Res. 8, 4 (1983), 538–548. Google ScholarDigital Library
- Daniele Micciancio and Panagiotis Voulgaris. 2013. A deterministic single exponential time algorithm for most lattice problems based on Voronoi cell computations. SIAM J. Comput. 42, 3 (2013), 1364–1391.Google ScholarDigital Library
- Oded Regev and Noah Stephens-Davidowitz. 2017. A Reverse Minkowksi Theorem. In STOC. Google ScholarDigital Library
- C.A. Rogers and G.C. Shephard. 1957.Google Scholar
- The difference body of a convex body. Arch. Math. 8 (1957), 220–233.Google Scholar
- C.-P. Schnorr. 1987.Google Scholar
- A hierarchy of polynomial time lattice basis reduction algorithms. Theoret. Comput. Sci. 53, 2-3 (1987), 201–224. 0304- 3975(87)90064- 8 Google ScholarDigital Library
- Uri Shapira and Barak Weiss. 2016. Stable lattices and the diagonal group. Journal of the European Mathematical Society 18, 8 (2016), 1753–1767.Google ScholarCross Ref
- Naftali Sommer, Meir Feder, and Ofir Shalvi. 2009.Google Scholar
- Finding the closest lattice point by iterative slicing. SIAM J. Discrete Math. 23, 2 (2009), 715–731. Abstract 1 Introduction 2 Our contributions 2.1 Techniques: Finding Dense Lattice Subspaces ReferencesGoogle Scholar
Index Terms
- On approximating the covering radius and finding dense lattice subspaces
Recommendations
Lattice Problems beyond Polynomial Time
STOC 2023: Proceedings of the 55th Annual ACM Symposium on Theory of ComputingWe study the complexity of lattice problems in a world where algorithms, reductions, and protocols can run in superpolynomial time. Specifically, we revisit four foundational results in this context—two protocols and two worst-case to average-case ...
Enumerative Lattice Algorithms in any Norm Via M-ellipsoid Coverings
FOCS '11: Proceedings of the 2011 IEEE 52nd Annual Symposium on Foundations of Computer ScienceWe give a novel algorithm for enumerating lattice points in any convex body, and give applications to several classic lattice problems, including the Shortest and Closest Vector Problems (SVP and CVP, respectively) and Integer Programming (IP). Our ...
The complexity of the covering radius problem
We initiate the study of the computational complexity of the covering radius problem for lattices, and approximation versions of the problem for both lattices and linear codes. We also investigate the computational complexity of the shortest linearly ...
Comments