Lower Bounds for Special Cases of Syntactic Multilinear ABPs
Algebraic Branching Programs (ABPs) are standard models for computing polynomials. Syntactic multilinear ABPs (smABPs) are restrictions of ABPs where every variable is allowed to occur at most once in every path from the start to terminal node. Proving lower bounds against syntactic multilinear ABPs remains a challenging open question in Algebraic Complexity Theory. The current best known bound is only quadratic [Alon,Kumar,Volk ECCC 2017].
In this article, we develop a new approach upper bounding the rank of the partial derivative matrix of syntactic multilinear ABPs: Convert the ABP to a syntactic multilinear formula with a super polynomial blow up in the size and then exploit the structural limitations of resulting formula to obtain a rank upper bound. Using this approach, we prove exponential lower bounds for special cases of smABPs and circuits namely, sum of Oblivious Read-Once ABPs, r-pass multilinear ABPs and sparse ROABPs. En route, we also prove super-polynomial lower bound for a special class of syntactic multilinear arithmetic circuits.
KeywordsComputational complexity Algebraic complexity theory Algebraic branching programs
- 1.Alon, N., Kumar, M., Volk, B.L.: An almost quadratic lower bound for syntactically multilinear arithmetic circuits. ECCC 24, 124 (2017). https://eccc.weizmann.ac.il/report/2017/124
- 3.Arvind, V., Raja, S.: Some lower bound results for set-multilinear arithmetic computations. Chicago J. Theoret. Comput. Sci. (2016). http://cjtcs.cs.uchicago.edu/articles/2016/6/contents.html
- 5.Chillara, S., Limaye, N., Srinivasan, S.: Small-depth multilinear formula lower bounds for iterated matrix multiplication, with applications. In: STACS (2018). http://arxiv.org/abs/1710.05481
- 6.Forbes, M.: Polynomial identity testing of read-once oblivious algebraic branching programs. Ph.D. thesis, Massachusetts Institute of Technology (2014)Google Scholar
- 8.Kayal, N., Nair, V., Saha, C.: Separation between read-once oblivious algebraic branching programs (ROABPs) and multilinear depth three circuits. In: STACS, pp. 46:1–46:15 (2016). https://doi.org/10.4230/LIPIcs.STACS.2016.46
- 10.Nisan, N.: Lower bounds for non-commutative computation (extended abstract). In: STOC, pp. 410–418 (1991). https://doi.org/10.1145/103418.103462
- 11.Ramya, C., Rao, B.V.R.: Sum of products of read-once formulas. In: FSTTCS, pp. 39:1–39:15 (2016). https://doi.org/10.4230/LIPIcs.FSTTCS.2016.39
- 15.Saptharishi, R.: A survey of lower bounds in arithmetic circuit complexity (2015). https://github.com/dasarpmar/lowerbounds-survey
- 16.Valiant, L.G.: Completeness classes in algebra. In: STOC, pp. 249–261 (1979). https://doi.org/10.1145/800135.804419