Schauder basis of ℓ^p
DefinitionThe unit vectors form an unconditional Schauder basis of ℓᵖ#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\), let \(1 \le p < \infty\) and let \(\iota\) be a type. This is the unconditional Schauder basis of \(\ell^p(\iota, \mathbb{K})\) whose \(i\)-th basis vector is the unit vector \(e_i\), equal to \(1\) at \(i\) and to \(0\) elsewhere, and whose \(i\)-th coordinate functional is the evaluation \(x \mapsto x_i\). In particular every \(x \in \ell^p(\iota, \mathbb{K})\) is the limit in norm of the finite partial sums \(\sum_{i \in A} x_i e_i\) along the finite sets \(A \subseteq \iota\) directed by inclusion.
MathlibExt.Analysis.Normed.Lp.SchauderBasis.lpUnconditionalSchauderBasis · 24 lines
noncomputable def lpUnconditionalSchauderBasis [DecidableEq ι] (hp : p ≠ ⊤) :
UnconditionalSchauderBasis ι 𝕜 (lp (fun _ : ι => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i)
(lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
· simp [hij, Pi.single_eq_of_ne]
expansion x := by
have h : HasSum (fun i : ι => lp.single (E := fun _ : ι => 𝕜) p i (x i)) x :=
lp.hasSum_single hp x
have hfun : (fun i : ι => (lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i) x • lp.single p i (1 : 𝕜))
= fun i : ι => lp.single (E := fun _ : ι => 𝕜) p i (x i) := by
funext i
show (x i) • lp.single (E := fun _ : ι => 𝕜) p i (1 : 𝕜)
= lp.single (E := fun _ : ι => 𝕜) p i (x i)
rw [← lp.single_smul, smul_eq_mul, mul_one]
rw [hfun]
exact h
Used by 6
DefinitionThe unit vectors form a Schauder basis of ℓᵖ#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\) and let \(1 \le p < \infty\). This is the Schauder basis of \(\ell^p(\mathbb{N}, \mathbb{K})\) whose \(i\)-th basis vector is the unit vector \(e_i\), equal to \(1\) at \(i\) and to \(0\) elsewhere, and whose \(i\)-th coordinate functional is the evaluation \(x \mapsto x_i\). In particular every \(x \in \ell^p(\mathbb{N}, \mathbb{K})\) is the limit in norm of its partial sums \(\sum_{i < n} x_i e_i\) as \(n \to \infty\).
MathlibExt.Analysis.Normed.Lp.SchauderBasis.lpSchauderBasis · 24 lines
noncomputable def lpSchauderBasis (hp : p ≠ ⊤) :
SchauderBasis 𝕜 (lp (fun _ : ℕ => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i)
(lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
· simp [hij, Pi.single_eq_of_ne]
expansion x := by
have h : HasSum (fun i : ℕ => lp.single (E := fun _ : ℕ => 𝕜) p i (x i)) x :=
lp.hasSum_single hp x
have hfun : (fun i : ℕ => (lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i) x • lp.single p i (1 : 𝕜))
= fun i : ℕ => lp.single (E := fun _ : ℕ => 𝕜) p i (x i) := by
funext i
show (x i) • lp.single (E := fun _ : ℕ => 𝕜) p i (1 : 𝕜)
= lp.single (E := fun _ : ℕ => 𝕜) p i (x i)
rw [← lp.single_smul, smul_eq_mul, mul_one]
rw [hfun]
exact h.mono_left SummationFilter.LeAtTop.le_atTop
Used by 9
TheoremCoordinate functionals of the unconditional ℓᵖ basis#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\), let \(1 \le p < \infty\) and let \(\iota\) be a type. For every \(i \in \iota\) and every \(x \in \ell^p(\iota, \mathbb{K})\), the \(i\)-th coordinate functional of the unconditional Schauder basis of \(\ell^p(\iota, \mathbb{K})\) by unit vectors sends \(x\) to its entry \(x_i\).
@[simp]
theorem lpUnconditionalSchauderBasis_coord_apply [DecidableEq ι] (hp : p ≠ ⊤)
(i : ι) (x : lp (fun _ : ι => 𝕜) p) :
(lpUnconditionalSchauderBasis (ι := ι) (𝕜 := 𝕜) (p := p) hp).coord i x = x i
noncomputable def lpUnconditionalSchauderBasis [DecidableEq ι] (hp : p ≠ ⊤) :
UnconditionalSchauderBasis ι 𝕜 (lp (fun _ : ι => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i)
(lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:52
Uses 1 · Used by 1
TheoremBasis vectors of the unconditional ℓᵖ basis#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\), let \(1 \le p < \infty\) and let \(\iota\) be a type. For every \(i \in \iota\), the \(i\)-th basis vector of the unconditional Schauder basis of \(\ell^p(\iota, \mathbb{K})\) by unit vectors is the unit vector \(e_i\), the family equal to \(1\) at \(i\) and to \(0\) elsewhere.
@[simp]
theorem lpUnconditionalSchauderBasis_apply [DecidableEq ι] (hp : p ≠ ⊤) (i : ι) :
(lpUnconditionalSchauderBasis (ι := ι) (𝕜 := 𝕜) (p := p) hp) i
= lp.single (E := fun _ : ι => 𝕜) p i (1 : 𝕜)
noncomputable def lpUnconditionalSchauderBasis [DecidableEq ι] (hp : p ≠ ⊤) :
UnconditionalSchauderBasis ι 𝕜 (lp (fun _ : ι => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i)
(lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:52
Uses 1 · Used by 1
TheoremFinite-set projections of the unconditional ℓᵖ basis#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\), let \(1 \le p < \infty\) and let \(\iota\) be a type. For every finite set \(A \subseteq \iota\) and every \(x \in \ell^p(\iota, \mathbb{K})\), the projection \(P_A\) of the unconditional Schauder basis of \(\ell^p(\iota, \mathbb{K})\) by unit vectors onto the basis vectors indexed by \(A\) satisfies \(P_A x = \sum_{i \in A} x_i e_i\), where \(e_i\) is the unit vector equal to \(1\) at \(i\) and to \(0\) elsewhere.
theorem lpUnconditionalSchauderBasis_proj_apply [DecidableEq ι] (hp : p ≠ ⊤)
(A : Finset ι) (x : lp (fun _ : ι => 𝕜) p) :
(lpUnconditionalSchauderBasis (ι := ι) (𝕜 := 𝕜) (p := p) hp).proj A x
= ∑ i ∈ A, lp.single (E := fun _ : ι => 𝕜) p i (x i)
noncomputable def lpUnconditionalSchauderBasis [DecidableEq ι] (hp : p ≠ ⊤) :
UnconditionalSchauderBasis ι 𝕜 (lp (fun _ : ι => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i)
(lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:52
Uses 3 · Used by 1
TheoremBasis projections of \(\ell^p\) are contractive#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\), let \(\iota\) be an index set and let \(1 \le p < \infty\). Let \(e_i\) (\(i \in \iota\)) be the unit vectors, which form an unconditional Schauder basis of \(\ell^p(\iota, \mathbb{K})\). For every finite set \(A \subseteq \iota\) and every \(x \in \ell^p(\iota, \mathbb{K})\), the basis projection \(P_A x = \sum_{i \in A} x_i e_i\) satisfies \(\|P_A x\|_p \le \|x\|_p\).
theorem norm_lpSchauderBasis_proj_apply_le [DecidableEq ι] (hp : p ≠ ⊤)
(A : Finset ι) (x : lp (fun _ : ι => 𝕜) p) :
‖(lpUnconditionalSchauderBasis (ι := ι) (𝕜 := 𝕜) (p := p) hp).proj A x‖ ≤ ‖x‖
noncomputable def lpUnconditionalSchauderBasis [DecidableEq ι] (hp : p ≠ ⊤) :
UnconditionalSchauderBasis ι 𝕜 (lp (fun _ : ι => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i)
(lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:52
TheoremThe \(\ell^p\) basis projections have operator norm at most \(1\)#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\), let \(\iota\) be an index set and let \(1 \le p < \infty\). Let \(e_i\) (\(i \in \iota\)) be the unit vectors, which form an unconditional Schauder basis of \(\ell^p(\iota, \mathbb{K})\). For every finite set \(A \subseteq \iota\), the basis projection \(P_A\), the bounded linear operator on \(\ell^p(\iota, \mathbb{K})\) given by \(P_A x = \sum_{i \in A} x_i e_i\), has operator norm \(\|P_A\| \le 1\).
theorem norm_lpSchauderBasis_proj_le [DecidableEq ι] (hp : p ≠ ⊤) (A : Finset ι) :
‖(lpUnconditionalSchauderBasis (ι := ι) (𝕜 := 𝕜) (p := p) hp).proj A‖ ≤ 1
noncomputable def lpUnconditionalSchauderBasis [DecidableEq ι] (hp : p ≠ ⊤) :
UnconditionalSchauderBasis ι 𝕜 (lp (fun _ : ι => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ι => 𝕜) p i)
(lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ι => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:52
Uses 2
TheoremCoordinate functionals of the ℓᵖ Schauder basis#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\) and let \(1 \le p < \infty\). For every \(i \in \mathbb{N}\) and every \(x \in \ell^p(\mathbb{N}, \mathbb{K})\), the \(i\)-th coordinate functional of the Schauder basis of \(\ell^p(\mathbb{N}, \mathbb{K})\) by unit vectors sends \(x\) to its \(i\)-th entry \(x_i\).
@[simp]
theorem lpSchauderBasis_coord_apply (hp : p ≠ ⊤)
(i : ℕ) (x : lp (fun _ : ℕ => 𝕜) p) :
(lpSchauderBasis (𝕜 := 𝕜) (p := p) hp).coord i x = x i
noncomputable def lpSchauderBasis (hp : p ≠ ⊤) :
SchauderBasis 𝕜 (lp (fun _ : ℕ => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i)
(lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:78
Uses 1 · Used by 1
TheoremBasis vectors of the ℓᵖ Schauder basis#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\) and let \(1 \le p < \infty\). For every \(i \in \mathbb{N}\), the \(i\)-th basis vector of the Schauder basis of \(\ell^p(\mathbb{N}, \mathbb{K})\) by unit vectors is the unit vector \(e_i\), the sequence equal to \(1\) at \(i\) and to \(0\) elsewhere.
@[simp]
theorem lpSchauderBasis_apply (hp : p ≠ ⊤) (i : ℕ) :
(lpSchauderBasis (𝕜 := 𝕜) (p := p) hp) i
= lp.single (E := fun _ : ℕ => 𝕜) p i (1 : 𝕜)
noncomputable def lpSchauderBasis (hp : p ≠ ⊤) :
SchauderBasis 𝕜 (lp (fun _ : ℕ => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i)
(lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:78
Uses 1 · Used by 1
TheoremThe ℓᵖ partial-sum projections are truncations#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\) and let \(1 \le p < \infty\). For every \(n \in \mathbb{N}\) and every \(x \in \ell^p(\mathbb{N}, \mathbb{K})\), the \(n\)-th partial-sum projection \(P_n\) of the Schauder basis of \(\ell^p(\mathbb{N}, \mathbb{K})\) by unit vectors satisfies \(P_n x = \sum_{i < n} x_i e_i\), where \(e_i\) is the unit vector equal to \(1\) at \(i\) and to \(0\) elsewhere; that is, \(P_n x\) is the truncation of \(x\) to its first \(n\) coordinates.
theorem lpSchauderBasis_proj_apply (hp : p ≠ ⊤)
(n : ℕ) (x : lp (fun _ : ℕ => 𝕜) p) :
(lpSchauderBasis (𝕜 := 𝕜) (p := p) hp).proj n x
= ∑ i ∈ Finset.range n, lp.single (E := fun _ : ℕ => 𝕜) p i (x i)
noncomputable def lpSchauderBasis (hp : p ≠ ⊤) :
SchauderBasis 𝕜 (lp (fun _ : ℕ => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i)
(lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:78
Uses 3 · Used by 1
TheoremThe ℓᵖ partial-sum projections are contractions#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\) and let \(1 \le p < \infty\). For every \(n \in \mathbb{N}\) and every \(x \in \ell^p(\mathbb{N}, \mathbb{K})\), the \(n\)-th partial-sum projection \(P_n\) of the Schauder basis of \(\ell^p(\mathbb{N}, \mathbb{K})\) by unit vectors satisfies \(\|P_n x\| \le \|x\|\).
theorem norm_lpSchauderBasis_classical_proj_apply_le (hp : p ≠ ⊤)
(n : ℕ) (x : lp (fun _ : ℕ => 𝕜) p) :
‖(lpSchauderBasis (𝕜 := 𝕜) (p := p) hp).proj n x‖ ≤ ‖x‖
noncomputable def lpSchauderBasis (hp : p ≠ ⊤) :
SchauderBasis 𝕜 (lp (fun _ : ℕ => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i)
(lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:78
Uses 3 · Used by 2
TheoremThe ℓᵖ partial-sum projections have norm at most one#✓fully proved
Let \(\mathbb{K}\) be \(\mathbb{R}\) or \(\mathbb{C}\) and let \(1 \le p < \infty\). For every \(n \in \mathbb{N}\), the \(n\)-th partial-sum projection of the Schauder basis of \(\ell^p(\mathbb{N}, \mathbb{K})\) by unit vectors has operator norm at most \(1\).
theorem norm_lpSchauderBasis_classical_proj_le (hp : p ≠ ⊤) (n : ℕ) :
‖(lpSchauderBasis (𝕜 := 𝕜) (p := p) hp).proj n‖ ≤ 1
noncomputable def lpSchauderBasis (hp : p ≠ ⊤) :
SchauderBasis 𝕜 (lp (fun _ : ℕ => 𝕜) p) where
basis i := lp.single p i 1
coord i := lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i
ortho i j := by
have h1 : (lp.evalCLM 𝕜 (fun _ : ℕ => 𝕜) p i)
(lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜))
= (lp.single (E := fun _ : ℕ => 𝕜) p j (1 : 𝕜)) i := rfl
rw [h1]
by_cases hij : i = j
· subst hij
simp [Pi.single_eq_same]
⋯
MathlibExt/Analysis/Normed/Lp/SchauderBasis.lean:78
Uses 2
Loading dependency explorer…
The dependency explorer needs JavaScript and its graph data.