LeanAide GPT-4 documentation

Module: Mathlib.Analysis.Convex.Caratheodory


eq_pos_convex_span_of_mem_convexHull

theorem eq_pos_convex_span_of_mem_convexHull : βˆ€ {π•œ : Type u_1} {E : Type u} [inst : LinearOrderedField π•œ] [inst_1 : AddCommGroup E] [inst_2 : Module π•œ E] {s : Set E} {x : E}, x ∈ (convexHull π•œ) s β†’ βˆƒ ΞΉ x_1 z w, Set.range z βŠ† s ∧ AffineIndependent π•œ z ∧ (βˆ€ (i : ΞΉ), 0 < w i) ∧ (Finset.univ.sum fun i => w i) = 1 ∧ (Finset.univ.sum fun i => w i β€’ z i) = x

This theorem is a more explicit version of `convexHull_eq_union`. It states that for any given type `π•œ` that is a linear ordered field, type `E` that is an additive commutative group and a module over `π•œ`, a set `s` of type `E`, and an element `x` of type `E`, if `x` is in the convex hull of `s` with respect to `π•œ`, then there exist elements `ΞΉ`, `x_1`, `z`, and `w` such that the range of `z` is a subset of `s`, `z` is affinely independent over `π•œ`, every element of `w` is positive, the sum of all elements of `w` is 1, and the sum of all elements of `w` multiplied by the corresponding elements of `z` is equal to `x`. Essentially, it is saying that any point in the convex hull can be expressed as a positive weighted sum of affinely independent points from the set, where the weights sum up to 1.

More concisely: Given a linear ordered field `π•œ`, an additive commutative group and `π•œ`-module `E`, a set `s` in `E`, and an element `x` in the convex hull of `s` with respect to `π•œ`, there exist `ΞΉ`, `x_1` in `s`, `z` in `s` affinely independent over `π•œ`, and `w` in `π•œ^(#s)` such that every component of `w` is positive, the sum of components of `w` is 1, and `x = βˆ‘(w * z)`, where `*` denotes scalar multiplication.

Caratheodory.mem_minCardFinsetOfMemConvexHull

theorem Caratheodory.mem_minCardFinsetOfMemConvexHull : βˆ€ {π•œ : Type u_1} {E : Type u} [inst : LinearOrderedField π•œ] [inst_1 : AddCommGroup E] [inst_2 : Module π•œ E] {s : Set E} {x : E} (hx : x ∈ (convexHull π•œ) s), x ∈ (convexHull π•œ) ↑(Caratheodory.minCardFinsetOfMemConvexHull hx)

The theorem `Caratheodory.mem_minCardFinsetOfMemConvexHull` states that for any linear ordered field `π•œ`, any additive commutative group `E` and module `π•œ E`, any set `s` of elements of type `E`, and any element `x` of type `E`, if `x` is an element of the convex hull of `s`, then `x` is also an element of the convex hull of the subset of `s` returned by the function `Caratheodory.minCardFinsetOfMemConvexHull`. This subset has the minimum cardinality among all subsets of `s` whose convex hull contains `x`. In other words, this theorem asserts that `x` is in the convex hull of the smallest possible subset of `s` that still includes `x` in its convex hull.

More concisely: For any linear ordered field π•œ, additive commutative group E and its module over π•œ, any set s of elements in E, and any x in E that belongs to the convex hull of s, x is also in the convex hull of the smallest subset of s with minimum cardinality guaranteeing x's inclusion.

Caratheodory.mem_convexHull_erase

theorem Caratheodory.mem_convexHull_erase : βˆ€ {π•œ : Type u_1} {E : Type u} [inst : LinearOrderedField π•œ] [inst_1 : AddCommGroup E] [inst_2 : Module π•œ E] [inst_3 : DecidableEq E] {t : Finset E}, Β¬AffineIndependent π•œ Subtype.val β†’ βˆ€ {x : E}, x ∈ (convexHull π•œ) ↑t β†’ βˆƒ y, x ∈ (convexHull π•œ) ↑(t.erase ↑y)

The theorem `Caratheodory.mem_convexHull_erase` states that for any type `π•œ`, which is a linearly ordered field, and any type `E`, which is an additive commutative group and a module over `π•œ` with a decidable equality, if `x` is an element of `E` that lies in the convex hull of a finset `t` of `E`, and the elements of `t` are not affinely independent, then there exists an element `y` in `t` such that `x` is in the convex hull of the set obtained by erasing `y` from `t`. In simpler terms, if a point is in the smallest convex set containing a non-affinely independent set, then this point is also in the smallest convex set containing a strict subset of the original set.

More concisely: If `x` is an element of a convex set `C` in a linearly ordered field `π•œ`, where `C` is generated by a finite, non-affinely independent set `t`, then `x` is also in the convex set generated by `t` without one of its elements.

convexHull_eq_union

theorem convexHull_eq_union : βˆ€ {π•œ : Type u_1} {E : Type u} [inst : LinearOrderedField π•œ] [inst_1 : AddCommGroup E] [inst_2 : Module π•œ E] {s : Set E}, (convexHull π•œ) s = ⋃ t, ⋃ (_ : ↑t βŠ† s), ⋃ (_ : AffineIndependent π•œ Subtype.val), (convexHull π•œ) ↑t

**CarathΓ©odory's convexity theorem** in the given context can be described as follows: For any set `s` of elements of type `E`, where `π•œ` is a linear ordered field and `E` is a group with an additional operation that forms a module over `π•œ`, the convex hull of `s` equals the union of convex hulls of all subsets `t` of `s` that are affinely independent. Here, the affine independence of a subset `t` means that no nontrivial weighted subtraction (where the sum of weights is zero) results in zero. The underlying element in the base type of a subset `t` is given by `Subtype.val`.

More concisely: The convex hull of a set in a linear ordered field with a group structure and module property is equal to the union of convex hulls of its affinely independent subsets.