Relax, but Don't be Too Lazy

@article{Hoeven2002RelaxBD,
  title={Relax, but Don't be Too Lazy},
  author={Joris van der Hoeven},
  journal={J. Symb. Comput.},
  year={2002},
  volume={34},
  pages={479-542}
}
  • J. Hoeven
  • Published 1 December 2002
  • Computer Science
  • J. Symb. Comput.
Assume that we wish to expand the product h = fg of two formal power series f and g. Classically, there are two types of algorithms to do this: zealous algorithms first expand f and g up to order n, multiply the results and truncate at order n. Lazy algorithms on the contrary compute the coefficients of f, g and h gradually and they perform no more computations than strictly necessary at each stage. In particular, at the moment we compute the coefficient hi of zi in h, only f0,..., fi and g0… 

Relaxed algorithms for p-adic numbers

This paper shows how to adapt the lazy and relaxed approaches to the context of p-adic numbers to the C++ library algebramix of Mathemagix, and shows significant speedups in the resolution ofp-adic functional equations when compared to the classical Newton iteration.

Improving Complexity Bounds for the Computation of Puiseux Series over Finite Fields

This paper reduces this bound to O~(d4+d2log q) using Hensel lifting and changes of variables in the Newton-Puiseux algorithm that give a better control of the number of steps.

Fast and Rigorous Computation of Special Functions to High Precision

This work gives new baby-step, giant-step algorithms for evaluation of linearly recurrent sequences involving an expensive parameter and for computing compositional inverses of power series and shows that isolated values of the integer partition function p(n) can be computed rigorously with softly optimal complexity by means of the Hardy-RamanujanRademacher formula and careful numerical evaluation.

Faster relaxed multiplication

  • J. Hoeven
  • Computer Science, Mathematics
    ISSAC
  • 2014
This paper will generalize the algorithm for relaxed power series multiplication to the cases when K is replaced by an effective ring of positive characteristic or by aneffective ring of characteristic zero, which is also torsion-free as a Z-module and comes with an additional algorithm for partial division by integers.

Fast composition of numeric power series ∗

Let f and g be two convergent power series in R[[z]] or C[[z]], whose first n terms are given numerically with a λ n-bit precision for a fixed constant λ > 0. Assuming that g0=0, we will show in this

On the Complexity and Parallel Implementation of Hensel's Lemma and Weierstrass Preparation

A complexity analysis is presented forHensel’s lemma, combined with repeated applications of Weierstrass preparation theorem, which allows for the factorization of polynomials with multivariate power series coefficients and shows the efficacy of this mixed parallel scheme.

Modular composition via factorization

How polynomial factorization may help modular composition is explored, and it is shown that compositions modulo a fixed modulus of composite degree are cheaper than general modular compositions.

Using approximate roots for irreducibility and equi-singularity issues in K[[x]][y]

We provide an irreducibility test in the ring K[[x]][y] whose complexity is quasi-linear with respect to the valuation of the discriminant, assuming the input polynomial F square-free and K a perfect

Relaxed mltiplication using the middle product

A new relaxed multiplication algorithm for the resolution of linear equations has the same asymptotic time complexity as previous algorithms, but the space overhead in the divide and conquer model and the constant factor in the F.F.T. model is improved.
...

References

SHOWING 1-10 OF 34 REFERENCES

CABAL: polynomial and power series algebra on a parallel computer

The work on designing or adapting algebra systems to exploit parallel hmdware has concentrated on trying the speed-up that can come from concurrent exploration of branches in search trees or overlapped computation in demanding algorithms, but this paper is almost exclusively concerned with the fact that large-scale parallel computers are frequently configured with very large amounts of primary memory.

Fast Algorithms for Manipulating Formal Power Series

This paper shows that the composition and reversion problems are equivalent (up to constant factors), and gives algorithms which require only order (n log n) ~/2 operations in many cases of practical importance.

An algorithm for the machine calculation of complex Fourier series

Good generalized these methods and gave elegant algorithms for which one class of applications is the calculation of Fourier series, applicable to certain problems in which one must multiply an N-vector by an N X N matrix which can be factored into m sparse matrices.

The Cycle Construction

The purpose in this note is to show that equality (0) follows directly from basic principles of combinatorial analysis 3], using elementary concepts of combinatorsics on words from Lothaire 4].

On bivariate Hensel lifting and its parallelization

An asymptotically faster algorithm is presented, which has a complexity of O((ln m)n 2 ln n) operations in the coeecient eld, using fast polynomial multiplication and O(n ln m) processors.

All Algebraic Functions Can Be Computed Fast

It is shown that the first N terms of an expansion of any algebraic function defined by an nth degree polynomial can be computed in O(n(M(N)) operations, while the classical method needs O(N sup n) operations.

Automatic Average-Case Analysis of Algorithm

On Short Multiplications and Divisions

  • T. Mulders
  • Computer Science, Mathematics
    Applicable Algebra in Engineering, Communication and Computing
  • 2000
A new way of choosing the size of the subproblems is proposed and the ideas used in the short multiplication algorithm are transferred to an algorithm for short divisions.

D-finite power series