Matrix.Isometry
defimported Lean
An isometry is a matrix A such that AAᴴ = 1. Compare with a unitary, which requires AAᴴ = AᴴA = 1. It is common to claim that, in a finite-dimensional vector space, a two-sided isometry (A.Isometry ∧ Aᴴ.Isometry) must be square and therefore unitary; this is does not work out so well here, since a Matrix m n R can be a two-sided isometry, but cannot be a unitary since the rows and columns are index by different labels.
Statement
set_option autoImplicit true in set_option relaxedAutoImplicit true in open scoped Matrix in variable {d d₂ d₃ R : Type*} in variable [Fintype d] [DecidableEq d] [Fintype d₂] [DecidableEq d₂] [Fintype d₃] [DecidableEq d₃] in variable [CommRing R] [StarRing R] in variable {𝕜 : Type*} [RCLike 𝕜] {A B : Matrix d d 𝕜} in def Matrix.Isometry (A : Matrix d d₂ R) : Prop
- Defined in
- QuantumInfo, `Matrix.Isometry`
- Lean proof
- By Alex Meiburg, from github.com/leanprover-community/physlib at
e411c6e89692(Copyright (c) 2025 Alex Meiburg. All rights reserved, Apache-2.0). Compiled here against leanprover/lean4:v4.33.1, Mathlib v4.33.1; written for leanprover/lean4:v4.34.1.Matrix.Isometry← Matrix.Isometry (QuantumInfo/ForMathlib/Isometry.lean:28): verbatim
- Verified
- 17 Oct 2026
- Axioms
- none listed
- Used by 9
- Matrix.isometry_preserves_normMatrix.isometry_preserves_norm
- Matrix.mem_unitaryGroup_iff_isometryMatrix.mem_unitaryGroup_iff_isometry
- Matrix.traceNorm_unitary_conjThe trace norm is invariant under unitary conjugation.
- Matrix.traceNorm_isometry_conjMatrix.traceNorm_isometry_conj
- Matrix.traceNorm_isometry_leftThe trace norm is invariant under left multiplication by an isometry.
- Matrix.traceNorm_isometry_rightThe trace norm is invariant under right multiplication by the adjoint of an isometry.
- Matrix.opNorm_isometryMatrix.opNorm_isometry
- Matrix.reindex_one_isometryMatrix.reindex_one_isometry
- Matrix.PosSemidef.traceNorm_eq_traceA positive semidefinite matrix has trace norm equal to its trace.
Lean source view module on GitHub
set_option autoImplicit true in set_option relaxedAutoImplicit true in open scoped Matrix in variable {d d₂ d₃ R : Type*} in variable [Fintype d] [DecidableEq d] [Fintype d₂] [DecidableEq d₂] [Fintype d₃] [DecidableEq d₃] in variable [CommRing R] [StarRing R] in variable {𝕜 : Type*} [RCLike 𝕜] {A B : Matrix d d 𝕜} in /-- An isometry is a matrix `A` such that `AAᴴ = 1`. Compare with a unitary, which requires `AAᴴ = AᴴA = 1`. It is common to claim that, in a finite-dimensional vector space, a two-sided isometry (`A.Isometry ∧ Aᴴ.Isometry`) must be square and therefore unitary; this is does not work out so well here, since a `Matrix m n R` can be a two-sided isometry, but cannot be a `unitary` since the rows and columns are index by different labels. -/ def Matrix.Isometry (A : Matrix d d₂ R) : Prop := Aᴴ * A = 1