Zulip Chat Archive

Stream: Is there code for X?

Topic: function equal to id of function (norm iterated_fderiv)


rtertr (Sonia) (Apr 24 2023 at 19:55):

Good evening everyone,

Is there a theorem for proving that these two norms are equal?
Or would is it not logically true, and they have to be written in a different way?

import tactic
import analysis.special_functions.exp_deriv
import analysis.calculus.iterated_deriv

noncomputable theory

namespace real

lemma inner_product_to_one_dimensional1 {i:} {a:} {x:  }: iterated_fderiv  i (λ (x :  ), x) (x)  = iterated_fderiv  i (λ (x :  ), x) x :=
begin
  simp only [real.norm_eq_abs],
  congr',
end

Filippo A. E. Nuccio (Apr 25 2023 at 13:35):

Well, what you are trying to prove is not true. For instance, in the special case i=1 after two rewrites one obtains

lemma inner_product_to_one_dimensional1 {a:} {x: }:
iterated_fderiv  1 (λ (x :  ), x) (x)  = iterated_fderiv  1 (λ (x :  ), x) x :=
begin
  rw norm_iterated_fderiv_eq_norm_iterated_deriv,
  rw iterated_deriv_one,
  rw norm_iterated_fderiv_eq_norm_iterated_deriv,
  rw iterated_deriv_one,
  simp,
end

with the goal ⊢ 1 = |deriv abs x| . Since the absolute value is not differentiable at 0, its derivative is defined to be 0 there, so the goal is actually ⊢ 1 = 0.


Last updated: Dec 20 2023 at 11:08 UTC