Abstract
We study projective dimension, a graph parameter, denoted by pd(G) for a bipartite graph G, introduced by Pudlák and Rödl (1992). For a Boolean function f (on n bits), Pudlák and Rödl associated a bipartite graph Gf and showed that size of the optimal branching program computing f, denoted by bpsize(f), is at least pd(Gf) (also denoted by pd(f)). Hence, proving lower bounds for pd(f) implies lower bounds for bpsize(f). Despite several attempts (Pudlák and Rödl (1992), Rónyai et al. (2000)), proving super-linear lower bounds for projective dimension of explicit families of graphs has remained elusive.
We observe that there exist a Boolean function f for which the gap between the pd(f) and bpsize(f)) is 2Ω(n). Motivated by the argument in Pudlák and Rödl (1992), we define two variants of projective dimension: projective dimension with intersection dimension 1, denoted by upd(f), and bitwise decomposable projective dimension, denoted by bitpdim(f). We show the following results:
(a) We observe that there exist a Boolean function f for which the gap between upd(f) and bpsize(f) is 2Ω(n). In contrast, we also show that the bitwise decomposable projective dimension characterizes size of the branching program up to a polynomial factor. That is, there exists a constant c > 0 and for any function f,
bitpdim(f)/6 ≤ bpsize(f) ≤ (bitpdim(f))c.
(b) We introduce a new candidate family of functions f for showing super-polynomial lower bounds for bitpdim(f). As our main result, for this family of functions, we demonstrate gaps between pd(f) and the above two new measures for f:
pd(f) = O(√n) upd(f) = Ω (n) bitpdim(f) = Ω (n1.5 / log n).
We adapt Nechiporuk’s techniques for our linear algebraic setting to prove the best-known bpsize lower bounds for bitpdim. Motivated by this linear algebraic setting of our main result, we derive exponential lower bounds for two restricted variants of pd(f) and upd(f) by observing that they are exactly equal to well-studied graph parameters—bipartite clique cover number and bipartite partition number, respectively.
- Romas Aleliunas, Richard M. Karp, Richard J. Lipton, László Lovász, and Charles Rackoff. 1979. Random walks, universal traversal sequences, and the complexity of maze problems. In Proceedings of the 20th Annual Symposium on Foundations of Computer Science. IEEE Computer Society, Los Alamitos, CA, 218--223. Google Scholar
Digital Library
- Eric Allender, Robert Beals, and Mitsunori Ogihara. 1999. The complexity of matrix rank and feasible systems of linear equations. Comput. Complex. 8, 2 (1999), 99--126. Google Scholar
Digital Library
- László Babai and Péter Frankl. 1992. Linear Algebra Methods in Combinatorics with Applications to Geometry and Computer Science. The University of Chicago, Chicago.Google Scholar
- Paul Beame, Nathan Grosshans, Pierre McKenzie, and Luc Segoufin. 2016. Nondeterminism and an abstract formulation of Nečiporuk’s lower bound method. Trans. Comput. Theory 9, 1 (2016), 5:1--5:34. Google Scholar
Digital Library
- Béla Bollobás. 2001. Random Graphs (2nd ed.). Cambridge University Press, Cambridge, UK.Google Scholar
- Philippe Delsarte. 1976. Association schemes and t-designs in regular semilattices. J. Combin. Theory A 20, 2 (Mar. 1976), 230--243.Google Scholar
Cross Ref
- Péter Frankl and Ronald L. Graham. 1985. Intersection theorems for vector spaces. Eur. J. Combin. 6, 2 (Jun. 1985), 183--187.Google Scholar
Cross Ref
- Péter Frankl and Richard M. Wilson. 1986. The Erdős-Ko-Rado theorem for vector spaces. J. Combin. Theory A 43, 2 (Nov. 1986), 228--236. Google Scholar
Digital Library
- Stasys Jukna. 2012. Boolean Function Complexity: Advances and Frontiers. Algorithms and Combinatorics, Vol. 27. Springer, New York, NY. Google Scholar
Digital Library
- Ralf Koetter and Frank R. Kschischang. 2008. Coding for errors and erasures in random network coding. IEEE Trans. Inf. Theory 54, 8 (2008), 3579--3591. Google Scholar
Digital Library
- Eyal Kushilevitz and Noam Nisan. 1997. Communication Complexity. Cambridge University Press, New York, NY, USA. Google Scholar
Digital Library
- Satyanarayana V. Lokam. 2009. Complexity lower bounds using linear algebra. Found. Trends Theor. Comput. Sci. 4, 182 (Jan. 2009), 1--155. Google Scholar
Digital Library
- Benjian Lv and Kaishun Wang. 2012. The eigenvalues of q-Kneser graphs. Discr. Math. 312, 6 (2012), 1144--1147. Google Scholar
Digital Library
- Michael Mitzenmacher and Eli Upfal. 2005. Probability and Computing: Randomized Algorithms and Probabilistic Analysis. Cambridge University Press, New York, NY. Google Scholar
Digital Library
- E. I. Nechiporuk. 1966. On a Boolean function. Dokl. Acad. Sci. USSR 164, 4 (1966), 765--766.Google Scholar
- Pavel Pudlák and Vojtěch Rödl. 1992. A combinatorial approach to complexity. Combinatorica 12, 2 (1992), 221--226. Google Scholar
Digital Library
- Pavel Pudlák and Vojtěch Rödl. 1994. Some combinatorial-algebraic problems from complexity theory. Discr. Math. 136, 1--3 (Dec. 1994), 253--279. Google Scholar
Digital Library
- Alexander A. Razborov. 1990. Applications of matrix methods to the theory of lower bounds in computational complexity. Combinatorica 10, 1 (1990), 81--93.Google Scholar
Cross Ref
- Omer Reingold. 2008. Undirected connectivity in log-space. J. ACM 55, 4 (2008), 17:1--17:24. Google Scholar
Digital Library
- Steven Roman. 2007. Advanced Linear Algebra (3rd ed.). Springer, New York, NY.Google Scholar
- Lajos Rónyai, László Babai, and Murali K. Ganapathy. 2002. On the number of zero-patterns of a sequence of polynomials. J. AMS 14 (2002), 2001.Google Scholar
- Heribert Vollmer. 1999. Introduction to Circuit Complexity: A Uniform Approach. Springer New York, NY. Google Scholar
Digital Library
Index Terms
Characterization and Lower Bounds for Branching Program Size using Projective Dimension
Recommendations
Pebbling, Entropy, and Branching Program Size Lower Bounds
We contribute to the program of proving lower bounds on the size of branching programs solving the Tree Evaluation Problem introduced by Cook et al. [2012]. Proving a superpolynomial lower bound for the size of nondeterministic thrifty branching ...
Kernelization Lower Bounds Through Colors and IDs
In parameterized complexity, each problem instance comes with a parameter k, and a parameterized problem is said to admit a polynomial kernel if there are polynomial time preprocessing rules that reduce the input instance to an instance with size ...
Lower bounds on kernelization
Preprocessing (data reduction or kernelization) to reduce instance size is one of the most commonly deployed heuristics in the implementation practice to tackle computationally hard problems. However, a systematic theoretical study of them has remained ...






Comments