CAP4KAM_nDOF: Computer-Assisted Proofs of existence of KAM tori in Hamiltonian systems with n (>=2) Degrees Of Freedom

In the folder you can produce by uncompressing the attached zipped
file (namely, that folder is called "CAP4KAM_nDOF"), you should find
everything you need in order to perform a complete computer-assisted
proof of existence of invariant tori for a Hamiltonian that satifies
three assumptions:
(i) it describes a (Hamiltonian) system with n>=2 degrees of
freedom and its canonical coordinates are n pairs of
action-angle variables;
(ii) it is close enough to a Kolmogorov normal form (so fulfilling
also both the non-resonance and the non-degeneracy conditions
usually adopted in the framework of KAM theory);
(iii) its expansion in Taylor series (with respect to the actions)
is finite, while its Fourier expansions (in the angles) can
be infinite.

The software included in the present folder "CAP4KAM_nDOF" is an
extension of a first public release that was a sort of supplementary
material of the paper [VL], i.e.,
Locatelli, Ugo (2021), “CAP4KAM2D: Computer-Assisted Proofs For
demonstrating the existence of 2-Dimensional KAM tori”, Mendeley
Data, V1, doi: 10.17632/jdx22ysh2s.1

The software included in the folder "CAP4KAM_nDOF" is designed to
be in a "easy-to-use" layout. Moreover, it is probably not too difficult
to be modified for people expert in programming (in C).
Everything about the files included in the folder "CAP4KAM_nDOF" is
widely described in the README.txt, that contains also careful
explanations that should be useful for running the codes, monitoring
the results, modifying the input files, etc.

REMARK: this is the second release of the software included in the
present folder "CAP4KAM_nDOF". With respect to the first release
some modifications have been introduced here. None of them has a
remarkable impact on the final results apart the correction of a
mistake in the expl_transf.c file. Indeed, in the previous version of such
a code, a functional norm of the generating function was computed in
such a way to be similar to the "max-norm" (for what concerns the
dependency on the actions), while the "l1-norm" has to be used
instead. This mistake has been removed in the present new release;
such a correction induces a worsening effect on the final
(rigorous) results that are provided by the Computer Assisted
Proofs. The results that are described in the folder "docs"
have been updated accordingly.

Eventual corrections or remarks about the software package included
in the folder "CAP4KAM_nDOF" are more than welcome and can be
sent to the author (Ugo Locatelli) at the following e-mail address:
[e-mail]