LeanAide GPT-4 documentation

Module: Mathlib.Analysis.Calculus.InverseFunctionTheorem.FDeriv



HasStrictFDerivAt.to_localInverse

theorem HasStrictFDerivAt.to_localInverse : βˆ€ {π•œ : Type u_1} [inst : NontriviallyNormedField π•œ] {E : Type u_2} [inst_1 : NormedAddCommGroup E] [inst_2 : NormedSpace π•œ E] {F : Type u_3} [inst_3 : NormedAddCommGroup F] [inst_4 : NormedSpace π•œ F] [inst_5 : CompleteSpace E] {f : E β†’ F} {f' : E ≃L[π•œ] F} {a : E} (hf : HasStrictFDerivAt f (↑f') a), HasStrictFDerivAt (HasStrictFDerivAt.localInverse f f' a hf) (↑f'.symm) (f a)

The theorem `HasStrictFDerivAt.to_localInverse` states that if a function `f` has an invertible derivative `f'` at a point `a`, in the sense of strict differentiability `(hf)`, then the inverse function (denoted as `hf.localInverse f`), computed using the derivative `f'`, has its own derivative `f'.symm` (the inverse of `f'`) at the point `f a`. This theorem is defined in the context of a certain type `π•œ` which has a nontrivial normed field structure, and two other types `E` and `F` which have normed additive commutative group and normed space structures over `π•œ`. `E` is also required to be a complete space.

More concisely: If a strict differentiable function `f : E β†’ F` with invertible derivative `f'` at `a ∈ E` has an invertible strict derivative, then the inverse function `hf.localInverse f` has the inverse derivative `f'.symm` at `f a`.

HasStrictFDerivAt.approximates_deriv_on_nhds

theorem HasStrictFDerivAt.approximates_deriv_on_nhds : βˆ€ {π•œ : Type u_1} [inst : NontriviallyNormedField π•œ] {E : Type u_2} [inst_1 : NormedAddCommGroup E] [inst_2 : NormedSpace π•œ E] {F : Type u_3} [inst_3 : NormedAddCommGroup F] [inst_4 : NormedSpace π•œ F] {f : E β†’ F} {f' : E β†’L[π•œ] F} {a : E}, HasStrictFDerivAt f f' a β†’ βˆ€ {c : NNReal}, Subsingleton E ∨ 0 < c β†’ βˆƒ s ∈ nhds a, ApproximatesLinearOn f f' s c

This theorem states that for a given function `f` from a normed additive commutative group `E` to another normed additive commutative group `F`, if `f` has a strict derivative `f'` at a point `a` and there is a nonnegative real number `c` such that `c` is greater than `0` or `E` is a subsingleton, then there exists a neighborhood `s` of `a` such that `f` approximates its derivative `f'` on this neighborhood with a constant `c`. This means that `f` behaves nearly linearly around `a` with a linear approximation given by `f'` and deviation measured by `c`. This theorem is foundational in the analysis of functions in normed spaces, underpinning the rigorous formalization of concepts such as continuity and differentiability.

More concisely: Given a continuous, norm-preserving function `f` from a normed additive commutative group `E` to another such group `F`, with strict derivative `f'` at `a ∈ E`, there exists a neighborhood `s` of `a` such that `||f(x) - f(a) - f'(a) * (x - a)|| ≀ c * ||x - a||` for all `x ∈ s`, where `c` is a positive constant.

open_map_of_strict_fderiv_equiv

theorem open_map_of_strict_fderiv_equiv : βˆ€ {π•œ : Type u_1} [inst : NontriviallyNormedField π•œ] {E : Type u_2} [inst_1 : NormedAddCommGroup E] [inst_2 : NormedSpace π•œ E] {F : Type u_3} [inst_3 : NormedAddCommGroup F] [inst_4 : NormedSpace π•œ F] [inst_5 : CompleteSpace E] {f : E β†’ F} {f' : E β†’ E ≃L[π•œ] F}, (βˆ€ (x : E), HasStrictFDerivAt f (↑(f' x)) x) β†’ IsOpenMap f

The theorem `open_map_of_strict_fderiv_equiv` states that if a function has an invertible strict derivative at all points, then the function is an open map. Here, an open map is defined as a function whose image of any open set remains open. In more formal terms, given a function `f` from a complete normed space `E` to another normed space `F`, both over a nontrivial normed field `π•œ`, and a function `f'` that assigns to each point in `E` an invertible bounded linear map from `E` to `F`, if `f` has an invertible strict derivative `f'` at every point in `E`, then `f` is an open map.

More concisely: If a function `f : E β†’ F` between complete normed spaces, with invertible and bounded strict derivatives `f'` at all points in `E`, then `f` is an open map.

HasStrictFDerivAt.approximates_deriv_on_open_nhds

theorem HasStrictFDerivAt.approximates_deriv_on_open_nhds : βˆ€ {π•œ : Type u_1} [inst : NontriviallyNormedField π•œ] {E : Type u_2} [inst_1 : NormedAddCommGroup E] [inst_2 : NormedSpace π•œ E] {F : Type u_3} [inst_3 : NormedAddCommGroup F] [inst_4 : NormedSpace π•œ F] {f : E β†’ F} {f' : E ≃L[π•œ] F} {a : E}, HasStrictFDerivAt f (↑f') a β†’ βˆƒ s, a ∈ s ∧ IsOpen s ∧ ApproximatesLinearOn f (↑f') s (‖↑f'.symmβ€–β‚Šβ»ΒΉ / 2)

This theorem states that for a function `f: E β†’ F` having a strict Frechet derivative `f'` at a point `a` in a normed space over a nontrivially normed field, there exists a set `s` such that `a` is in `s`, `s` is open in the topological space, and `f` approximates linearly on `s` with error term bounded by half the inverse of the norm of the continuous linear inverse of `f'`. This essentially means that around every point where the function `f` has a derivative, we can find an open set where the function behaves nearly linearly, with a precise bound given on the error.

More concisely: Given a function `f: E β†’ F` with a strict Frechet derivative `f'` at a point `a` in a normed space over a nontrivially normed field, there exists an open set `s` containing `a` such that `f` approximates `f(x) β‰ˆ f(a) + f'(a)(x - a)` for all `x ∈ s`, with an error bound `||(x - a) - f'(a)⁻¹(f(x) - f(a) - f'(a)(x - a))|| ≀ ||f'(a)⁻¹||/2`.

HasStrictFDerivAt.mem_toPartialHomeomorph_source

theorem HasStrictFDerivAt.mem_toPartialHomeomorph_source : βˆ€ {π•œ : Type u_1} [inst : NontriviallyNormedField π•œ] {E : Type u_2} [inst_1 : NormedAddCommGroup E] [inst_2 : NormedSpace π•œ E] {F : Type u_3} [inst_3 : NormedAddCommGroup F] [inst_4 : NormedSpace π•œ F] [inst_5 : CompleteSpace E] {f : E β†’ F} {f' : E ≃L[π•œ] F} {a : E} (hf : HasStrictFDerivAt f (↑f') a), a ∈ (HasStrictFDerivAt.toPartialHomeomorph f hf).source

The theorem `HasStrictFDerivAt.mem_toPartialHomeomorph_source` states that for a function `f` from a normed add commutative group `E` to another normed add commutative group `F`, both over a nontrivially normed field `π•œ`, if this function has a strict Frechet derivative at a point `a` in `E`, then `a` belongs to the source of the partial homeomorphism associated with the strict Frechet derivative.

More concisely: If `f : E β†’ F` is a function between normed add commutative groups over a nontrivially normed field with a strict Frechet derivative at `a ∈ E`, then `a` is in the domain of the partial homeomorphism defined by the strict Frechet derivative.

HasStrictFDerivAt.to_local_left_inverse

theorem HasStrictFDerivAt.to_local_left_inverse : βˆ€ {π•œ : Type u_1} [inst : NontriviallyNormedField π•œ] {E : Type u_2} [inst_1 : NormedAddCommGroup E] [inst_2 : NormedSpace π•œ E] {F : Type u_3} [inst_3 : NormedAddCommGroup F] [inst_4 : NormedSpace π•œ F] [inst_5 : CompleteSpace E] {f : E β†’ F} {f' : E ≃L[π•œ] F} {a : E}, HasStrictFDerivAt f (↑f') a β†’ βˆ€ {g : F β†’ E}, (βˆ€αΆ  (x : E) in nhds a, g (f x) = x) β†’ HasStrictFDerivAt g (↑f'.symm) (f a)

This theorem states that if we have a function `f : E β†’ F`, with an invertible derivative `f'` at the point `a` in the sense of strict differentiability, and another function `g` that undoes the effect of `f` in the neighborhood of `a` - i.e., `g (f x) = x` - then `g` has a derivative at the point `f a` that is the inverse of `f'`. The theorem assumes that the domain `E` is a complete space. A variant of this theorem is also available that doesn't require `E` to be complete, but instead requires `f (g y) = y` and the continuity of `g` at `f a`.

More concisely: If `f : E β†’ F` is a strict differentiable function with invertible derivative `f'` at `a ∈ E` and there exists a function `g : F β†’ E` such that `g (f x) = x` in a neighborhood of `f a`, then `g` has a derivative at `f a` and `(g ')(f a) = (f ')⁻¹`.

HasStrictFDerivAt.image_mem_toPartialHomeomorph_target

theorem HasStrictFDerivAt.image_mem_toPartialHomeomorph_target : βˆ€ {π•œ : Type u_1} [inst : NontriviallyNormedField π•œ] {E : Type u_2} [inst_1 : NormedAddCommGroup E] [inst_2 : NormedSpace π•œ E] {F : Type u_3} [inst_3 : NormedAddCommGroup F] [inst_4 : NormedSpace π•œ F] [inst_5 : CompleteSpace E] {f : E β†’ F} {f' : E ≃L[π•œ] F} {a : E} (hf : HasStrictFDerivAt f (↑f') a), f a ∈ (HasStrictFDerivAt.toPartialHomeomorph f hf).target

This theorem states that for any non-trivial normed field `π•œ`, normed additive commutative groups `E` and `F` equipped with a normed space structure over `π•œ`, where `E` is complete, if a function `f : E β†’ F` has a strict FrΓ©chet derivative at a point `a` in `E`, represented by the linear map `f' : E ≃L[π•œ] F`, then the image of `a` under `f` is within the target set of the partial homeomorphism induced by the strict derivative of `f` at `a`.

More concisely: If `f : E β†’ F` is a strict differentiable function from a complete normed additive commutative group `E` to a normed space `F` over a non-trivial normed field `π•œ`, with strict derivative `f' : E ≃L[π•œ] F` at a point `a` in `E`, then `f(a)` lies in the image of the partial homeomorphism induced by `f'` at `a`.

HasStrictFDerivAt.map_nhds_eq_of_equiv

theorem HasStrictFDerivAt.map_nhds_eq_of_equiv : βˆ€ {π•œ : Type u_1} [inst : NontriviallyNormedField π•œ] {E : Type u_2} [inst_1 : NormedAddCommGroup E] [inst_2 : NormedSpace π•œ E] {F : Type u_3} [inst_3 : NormedAddCommGroup F] [inst_4 : NormedSpace π•œ F] [inst_5 : CompleteSpace E] {f : E β†’ F} {f' : E ≃L[π•œ] F} {a : E}, HasStrictFDerivAt f (↑f') a β†’ Filter.map f (nhds a) = nhds (f a)

This theorem states that for a given strict Frechet derivative at a point `a` in a complete normed additive commutative group `E` over a nontrivially normed field `π•œ`, mapped to another normed additive commutative group `F` under a continuous linear equivalence `f'`, if the function `f` has a strict Frechet derivative `f'` at `a`, then the forward map of the neighborhood filter at `a` under `f` is equal to the neighborhood filter at `f(a)`. In simpler terms, this theorem provides a condition under which the derivative of a function at a point determines the local behavior of the function around that point in the sense of filters.

More concisely: If `f : E β†’ F` is a continuous linear map between complete normed additive commutative groups with `f'` its strict Frechet derivative at `a ∈ E`, then the forward neighborhood filter of `a` under `f` equals the neighborhood filter of `f(a)`.