Abstract
In this paper we describe a new technique for obtaining lower bounds on restricted classes of non-monotone arithmetic circuits. The heart of this technique is a complexity measure for multivariate polynomials, based on the linear span of their partial derivatives. We use the technique to obtain new lower bounds for computing symmetric polynomials (that hold over fields of characteristic zero) and iterated matrix products (that hold for all fields).
Similar content being viewed by others
References
N. Alon, J. Spencer., and P. Erdös,The probabilistic method. Wiley, 1991.
M. Ben-Or, Private communication.
J. von zur Gathen, Algebraic complexity theory.Annual Review of Computer Science 3 (1988), 317–347.
G. H. Gottlieb, A certain class of incidence matrices. InProc. AMS 17, 1966, 1233–1237.
N. Nisan, Lower bounds for non-commutative computation. InSTOC 1991, 1991, 410–418.
E. Shamir andM. Snir, On the depth complexity of formulas.Math. Systems theory 13 (1980), 301–322.
Roman Smolensky, On interpolation by analytic functions with special properties and some weak lower bounds on the size of circuits with symmetric gates. In31st FOCS, 1990, 628–631.
V. Strassen Vermeidung von Divisionen.J. reine u. angew. Math. 264 (1973), 182–202.
P. Tiwari and M. Tompa, A direct version of Shamir and Snir's lower bounds on monotone circuit depth.Information Processing Letters (1994), p. 49.
L. Valiant, Negation can be exponentially powerful.TCS 12 11 (1980), 303–314.
L. Valiant, S. Skyum, S. Berkowitz, andC. Rackoff, Fast parallel computation of polynomials using few processors.SIAM J. Comput. 12 (1983), 641–644.
Author information
Authors and Affiliations
Additional information
Dedicated to the memory of Roman Smolensky
Rights and permissions
About this article
Cite this article
Nisan, N., Wigderson, A. Lower bounds on arithmetic circuits via partial derivatives. Comput Complexity 6, 217–234 (1996). https://doi.org/10.1007/BF01294256
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF01294256