Finite-type invariants of three-manifolds and the dimension subgroup problem
Gwenael Massuyeau
Abstract
For a certain class of compact oriented 3-manifolds, M. Goussarov and K. Habiro have conjectured that the information carried by finite-type invariants should be characterized in terms of ``cut-and-paste'' operations defined by the lower central series of the Torelli group of a surface. In this paper, we observe that this is a variation of a classical problem in group theory, namely the ``dimension subgroup problem.'' This viewpoint allows us to prove, by purely algebraic methods, an analogue of the Goussarov-Habiro conjecture for finite-type invariants with values in a fixed field. We deduce that their original conjecture is true at least in a weaker form.
Create a lesson
Related papers
Mapping class groups have a unique Polish group structure
Tyrone Ghaswala, Sumun Iyer, Robert Alonzo Lyman et al.
Around the Andreadakis-Johnson filtration
Yusuke Kuno
Mapping class groups of simply connected closed spin 5-manifolds with no 2- and 3-torsion in homology
Huize Jin
A note on surfaces with large systoles
Yifei Cai
Real ideal points, Conway spheres, and left-orderable Dehn fillings
Yi Wang
All once-extended 3D TQFTs are Reshetikhin--Turaev theories
Glen Lim