[Paper Review] A Direct Ultrametric Approach to Additive Complexity and the Shub-Smale Tau Conjecture

J. Maurice Rojas|ArXiv.org|Apr 7, 2003

Polynomial and algebraic computation9 references3 citations

TL;DR

This paper presents a novel ultrametric approach to additive complexity using p-adic analysis, proving that a univariate polynomial of additive complexity $ s $ has at most $ O(e^{s\log s}) $ roots in $ \mathbb{Q}_2 $, significantly improving prior bounds. It establishes weaker versions of the Shub-Smale $ \tau $-conjecture and shows that the truth of the $ p $-adic Digit Conjecture for any prime $ p $ implies $ \mathbf{P}_{\mathbb{C}} \neq \mathbf{NP}_{\mathbb{C}} $, advancing a number-theoretic route to resolving the BSS model's $ \mathbf{P} \stackrel{?}{=} \mathbf{NP} $ question.

ABSTRACT

The Shub-Smale Tau Conjecture is a hypothesis relating the number of integral roots of a polynomial f in one variable and the Straight-Line Program (SLP) complexity of f. A consequence of the truth of this conjecture is that, for the Blum-Shub-Smale model over the complex numbers, P differs from NP. We prove two weak versions of the Tau Conjecture and in so doing show that the Tau Conjecture follows from an even more plausible hypothesis. Our results follow from a new p-adic analogue of earlier work relating real algebraic geometry to additive complexity. For instance, we can show that a nonzero univariate polynomial of additive complexity s can have no more than 15+s^3(s+1)(7.5)^s s! =O(e^{s\log s}) roots in the 2-adic rational numbers Q_2, thus dramatically improving an earlier result of the author. This immediately implies the same bound on the number of ordinary rational roots, whereas the best previous upper bound via earlier techniques from real algebraic geometry was a quantity in Omega((22.6)^{s^2}). This paper presents another step in the author's program of establishing an algorithmic arithmetic version of fewnomial theory.

Motivation & Objective

• To establish sharper upper bounds on the number of rational roots of polynomials using ultrametric methods.
• To provide evidence for the Shub-Smale $ \tau $-conjecture by proving weak variants via $ p $-adic analysis.
• To connect additive complexity to the $ \mathbf{P}_{\mathbb{C}} \neq \mathbf{NP}_{\mathbb{C}} $ conjecture through number-theoretic hypotheses.
• To develop an algorithmic arithmetic version of fewnomial theory using $ p $-adic geometry.

Proposed method

• Utilizes $ p $-adic valuation and ultrametric geometry to analyze root distributions of polynomials over $ \mathbb{Q}_p $.
• Applies the $ p $-adic Complex Fewnomial Theorem to bound the number of geometrically isolated roots in $ \mathbb{C}_p^{s+1} $.
• Defines additive complexity $ \sigma_R(f) $ as the minimal number of operations needed to express $ f $ via nested sums and products over a ring $ R $.
• Reduces the root-counting problem to counting isolated solutions of a system of polynomial equations derived from the SLP representation.
• Uses rescaling and valuation constraints to restrict root locations within $ p $-adic balls of radius $ 1/p^r $.
• Applies combinatorial bounds on monomial structures to control the number of possible root configurations.

Experimental results

Research questions

• RQ1 Can the Shub-Smale $ \tau $-conjecture be approached via $ p $-adic analogues of real algebraic geometry?
• RQ2 What is the maximum number of $ \mathbb{Q}_2 $-rational roots a univariate polynomial of additive complexity $ s $ can have?
• RQ3 Does the truth of the $ p $-adic Digit Conjecture for any prime $ p $ imply $ \mathbf{P}_{\mathbb{C}} \neq \mathbf{NP}_{\mathbb{C}} $?
• RQ4 Can additive complexity be bounded using ultrametric techniques to yield tighter root count estimates than classical fewnomial theory?
• RQ5 How does the structure of straight-line programs relate to the distribution of roots in $ p $-adic fields?

Key findings

• A nonzero univariate polynomial of additive complexity $ s $ has at most $ 1 + s^3(s+1)(7.5)^s s! = O(e^{s\log s}) $ roots in $ \mathbb{Q}_2 $, a dramatic improvement over the previous $ \Omega((22.6)^{s^2}) $ bound.
• The same bound applies to ordinary rational roots, providing a new, significantly tighter upper bound on the number of integral roots.
• The $ p $-adic Digit Conjecture for any fixed prime $ p $ implies $ \mathbf{P}_{\mathbb{C}} \neq \mathbf{NP}_{\mathbb{C}} $, and thus implies the full $ \tau $-conjecture.
• The number of $ p $-adic roots with $ |x - 1|_p \leq 1/p^r $ is bounded by a function involving $ C_p(k,1,r) $, which controls root multiplicity via fewnomial-type bounds.
• The proof establishes that $ N_p(s) \geq s $, showing that additive complexity $ s $ can support at least $ s $ distinct $ p $-adic valuations of roots.
• The method yields a constructive link between SLP complexity and the geometry of roots in $ \mathbb{C}_p^{s+1} $, via isomorphism of quotient rings $ \mathbb{C}_p[x]/\langle f \rangle \cong \mathbb{C}_p[X_1,\ldots,X_{s+1}]/\langle G \rangle $.