The Minkowski functional, normed field version #
In this file we define (egauge π s Β·)
to be the Minkowski functional (gauge) of the set s
in a topological vector space E over a normed field π,
as a function E β ββ₯0β.
It is defined as the infimum of the norms of c : π such that x β c β’ s.
In particular, for π = ββ₯0 this definition gives an ββ₯0β-valued version of gauge
defined in Mathlib/Analysis/Convex/Gauge.lean.
This definition can be used to generalize the notion of FrΓ©chet derivative to maps between topological vector spaces without norms.
Currently, we can't reuse results about egauge for gauge,
because we lack a theory of normed semifields.
The Minkowski functional for vector spaces over normed fields.
Given a set s in a vector space over a normed field π,
egauge s is the functional which sends x : E
to the infimum of βcββ over c such that x belongs to s scaled by c.
The definition only requires π to have a NNNorm instance
and (Β· β’ Β·) : π β E β E to be defined.
This way the definition applies, e.g., to π = ββ₯0.
For π = ββ₯0, the function is equal (up to conversion to β)
to the usual Minkowski functional defined in gauge.
Instances For
Alias of the reverse direction of egauge_zero_left_eq_top.
If c β’ x β s and c β 0, then egauge π s x is at most `((βcβββ»ΒΉ : ββ₯0) : ββ₯0β).
See also egauge_le_of_smul_mem.
If c β’ x β s, then egauge π s x is at most `(βcββ : ββ₯0β)β»ΒΉ.
See also egauge_le_of_smul_mem_of_ne.