The approximating sequence of Henkel's open mapping theorem #
TauCeti/Topology/Algebra/OpenMapping/Basic.lean proves one step of Henkel's approximation: a
point in
closure (f '' U) is brought inside closure (f '' V) by subtracting the image of some x ∈ U.
Iterating that step down a sequence of sets produces a sequence of approximants whose residuals
shrink, and this file builds it.
Nothing here converges anything: the sequence exists in any topological additive group, with no
completeness and no first countability. Summing it — which is what finally removes the closure
from f '' U — is a separate step, and the two hypotheses enter there separately:
completeness sums the resulting null sequence, while countable generation of 𝓝 0 is what
produces the sequence of subgroups to sum along. TauCeti.mem_image_of_mem_closure_image takes
that sequence as an explicit HasAntitoneBasis hypothesis rather than synthesising it, so it
assumes completeness but no first-countability instance.
Main results #
TauCeti.exists_seq_mem_and_sub_sum_mem: the approximating sequence, with its partial sums driving the residual intoclosure (f '' V (n + 1))at every stage.
References #
- L. Henkel, An Open Mapping Theorem for rings which have a zero sequence of units, arXiv:1407.5647.
Henkel's approximating sequence. If each closure (f '' V (n + 1)) is a neighbourhood of
zero and y lies in closure (f '' V 0), there are approximants x n ∈ V n whose partial sums
drive the residual down the sequence: y - f (x 0 + ⋯ + x n) lies in closure (f '' V (n + 1)).
No completeness or first countability is used: those are what let the sequence be summed, which
is what finally removes the closure from f '' U, and neither is needed to obtain the
approximants.