Abstract
In this paper, we consider the computation of the volume of an n-dimensional crosspolytope truncated by a halfspace. Since a crosspolytope has exponentially many facets, we cannot efficiently compute the volume by dividing the truncated crosspolytope into simplices. We show an \(O(n^6)\) time algorithm for the computation of the volume. This makes a contrast to the 0−1 knapsack polytope, whose volume is \(\#P\)-hard to compute. The paper is interested in the computation of the volume of the truncated crosspolytope because we conjecture the following question may have an affirmative answer: Does the existence of a polynomial time algorithm for the computation of the volume of a polytope K imply the same for K’s geometric dual? We give one example where the answer is yes.
This research was supported by research grant of Information Sciences Institute of Senshu University.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Though the number M of the simplices may be exponentially large with respect to n, it does not matter because, in the later part of the paper, we show another way to compute \(F_K(\varvec{a},b)\) efficiently for the case where K is a crosspolytope.
References
Ando, E.: An FPTAS for computing the distribution function of the longest path length in DAGs with uniformly distributed edge lengths. In: Poon, S.-H., Rahman, M.S., Yen, H.-C. (eds.) WALCOM 2017. LNCS, vol. 10167, pp. 421–432. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-53925-6_33
Ando, E., Kijima, S.: An FPTAS for the volume of a \({\cal{V}}\)-polytopes—it is hard to compute the volume of the intersection of two cross-polytopes. arXiv:1607.06173
Ando, E., Kijima, S.: An FPTAS for the volume computation of 0-1 knapsack polytopes based on approximate convolution. Algorithmica 76(4), 1245–1263 (2016)
Ando, E., Kijima, S.: An FPTAS for the volume of some \(\cal{V}\)-polytopes—it is hard to compute the volume of the intersection of two cross-polytopes. In: Cao, Y., Chen, J. (eds.) COCOON 2017. LNCS, vol. 10392, pp. 13–24. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-62389-4_2
Bárány, I., Füredi, Z.: Computing the volume is difficult. Discrete Comput. Geom. 2, 319–326 (1987)
Cousins, B., Vempala, S.: Bypassing KLS: Gaussian cooling and an \(O^\ast (n^3)\) volume algorithm. In: Proceedings of the STOC 2015, pp. 539–548 (2015)
Dyer, M., Frieze, A., Kannan, R.: A random polynomial-time algorithm for approximating the volume of convex bodies. J. Assoc. Comput. Mach. 38(1), 1–17 (1991)
Dyer, M., Gritzmann, P., Hufnagel, A.: On the complexity of computing mixed volumes. SIAM J. Comput. 27(2), 356–400 (1998)
Dyer, M., Frieze, A.: On the complexity of computing the volume of a polyhedron. SIAM J. Comput. 17(5), 967–974 (1988)
Elekes, G.: A geometric inequality and the complexity of computing volume. Discrete Comput. Geom. 1, 289–292 (1986)
Gopalan, P., Klivans, A., Meka, R., Štefankovič, D., Vempala, S., Vigoda, E.: An FPTAS for #knapsack and related counting problems. In: Proceedings of FOCS 2011, pp. 817–826 (2011)
Khachiyan, L.: The problem of computing the volume of polytopes is \(\#P\)-hard. Uspekhi Mat. Nauk. 44, 199–200 (1989)
Khachiyan, L.: Complexity of polytope volume computation. In: Pach, J. (ed.) New Trends in Discrete and Computational Geometry. AC, vol. 10, pp. 91–101. Springer, Heidelberg (1993). https://doi.org/10.1007/978-3-642-58043-7_5
Li, J., Shi, T.: A fully polynomial-time approximation scheme for approximating a sum of random variables. Oper. Res. Lett. 42, 197–202 (2014)
Lovász, L., Vempala, S.: Simulated annealing in convex bodies and an \(O^\ast (n^4)\) volume algorithm. J. Comput. Syst. Sci. 72, 392–417 (2006)
Matous̆ek, J.: Lectures on Discrete Geometry. Graduate Texts in Mathematics. Springer, New York (2002). https://doi.org/10.1007/978-1-4613-0039-7
Štefankovič, D., Vempala, S., Vigoda, E.: A deterministic polynomial-time approximation scheme for counting knapsack solutions. SIAM J. Comput. 41(2), 356–366 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix Supplemental Proof
Appendix Supplemental Proof
To make this paper self-contained, we prove the following proposition which is used in the proof of Proposition 3.
Proposition 4
Let \(P\subseteq \mathbb {R}^n\) be a convex n-dimensional polytope. Then there exists a set of m simplices \(S_1,\dots ,S_m\) satisfying the following three conditions:
-
1.
\(P=\bigcup _{i=1,\dots ,m} S_i\);
-
2.
any vertex of the simplices \(S_1,\dots ,S_m\) is a vertex of P;
-
3.
\(\mathrm {Vol}(S_i\cap S_j)=0\) for any \(1\le i < j \le m\).
Proof
The proof is the induction on n. As for the base case, we consider the case \(n=1\). In this case, P is always a bounded interval, which is a simplex. Therefore, the proposition holds for the base case.
We proceed to the induction step. We assume that we have the claims of the proposition in case \(n=k\). Then, in case \(n=k+1\), we have that any facet of P can be divided into a set of k-dimensional simplices, satisfying the three conditions of the claim. Let \(S'_1,\dots ,S'_M\) be the k-dimensional simplices obtained by dividing the P’s facets satisfying the three conditions for each facet. Let \(\varvec{v}\) be one vertex of P. Then, we obtain the \((k+1)\)-dimensional simplices as the convex hulls \(S_i=\mathrm {conv}(S'_i\cup \{\varvec{v}\})\) for \(i=1,\dots , M\).
As for the first condition of the proposition, we show that for any internal point \(\varvec{p}\in P\), there exists a point \(\varvec{q}\in S'_i\) for some \(1\le i \le M\) such that \(\varvec{p}\in \mathrm {conv}(\{\varvec{v},\varvec{q}\})\). We consider a point given by \(\varvec{r}(t)=t(\varvec{p}-\varvec{v})+\varvec{v}\), where \(t>0\). Since P is bounded, we have that \(\varvec{r}(t)\) is on a facet F of P for some \(t>1\). We have \(t>1\) since \(\varvec{p}\) is an internal point of P. Since each facet F can be divided into simplices, \(\varvec{q}=\varvec{r}(t)\) is in one of these simplices.
Since the second condition of the claim clearly holds for \(S_1,\dots ,S_M\) by definition, we proceed to the proof of the third condition. That is, \(\mathrm {Vol}(S_i\cap S_j)=0\) for any \(1\le i < j \le M\). Let \(\varvec{p}\in S_i\cap S_j\). We consider the point \(\varvec{r}(t)=t(\varvec{p}-\varvec{v})+\varvec{v}\) as in the above for \(t>1\). Let \(t_0\) be the value of t such that \(\varvec{r}(t_0)\) is on the surface of P. Since \(S_i=\mathrm {conv}(S'_i\cup \{\varvec{v}\})\) and \(S_j=\mathrm {conv}(S'_j\cup \{\varvec{v}\})\), we have that \(\varvec{r}(t_0)\in S'_i\cap S'_j\). Since the k-dimensional volume of \(S'_i\cap S'_j\) is 0 by the assumption, we have that \(\mathrm {Vol}(\mathrm {conv}(\{\varvec{v}\}\cup (S'_i\cap S'_j)))=0\), which shows the claim. \(\square \)
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Ando, E., Tsuchiya, S. (2019). The Volume of a Crosspolytope Truncated by a Halfspace. In: Gopal, T., Watada, J. (eds) Theory and Applications of Models of Computation. TAMC 2019. Lecture Notes in Computer Science(), vol 11436. Springer, Cham. https://doi.org/10.1007/978-3-030-14812-6_2
Download citation
DOI: https://doi.org/10.1007/978-3-030-14812-6_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-14811-9
Online ISBN: 978-3-030-14812-6
eBook Packages: Computer ScienceComputer Science (R0)