In mathematics, and in particular linear algebra, the Moore–Penrose inverse
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/86b8a304-b510-4a3e-a01d-7c1f593dda2c/b380a5ff4e2d7d22a0dc1aea46e7ecba61f95fe6
of a matrix
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/fc0b39c1-5f5b-4ef2-a41c-c7fd2e0be5c1/7daff47fa58cdfd29dc333def748ff5fa4c923e3
is the most widely known generalization of the inverse matrix.[1][2][3][4] It was independently described by E. H. Moore[5] in 1920, Arne Bjerhammar[6] in 1951, and Roger Penrose[7] in 1955. Earlier, Erik Ivar Fredholm had introduced the concept of a pseudoinverse of integral operators in 1903. When referring to a matrix, the term pseudoinverse, without further specification, is often used to indicate the Moore–Penrose inverse. The term generalized inverse is sometimes used as a synonym for pseudoinverse.
A common use of the pseudoinverse is to compute a "best fit" (least squares) solution to a system of linear equations that lacks a unique solution (see below under § Applications). Another use is to find the minimum (Euclidean) norm solution to a system of linear equations with multiple solutions. The pseudoinverse facilitates the statement and proof of results in linear algebra.
The pseudoinverse is defined and unique for all matrices whose entries are real or complex numbers. It can be computed using the singular value decomposition.
In the following discussion, the following conventions are adopted.
will denote one of the fields of real or complex numbers, denoted , , respectively. The vector space of matrices over is denoted by .
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/2818ea78-7b84-4952-a806-d50bff20ec5b/1848c435e64864e9ad4efa7e46bd6bc900c35c99
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/44010d1d-601a-4425-ae8d-eade47bcfa0e/786849c765da7a84dbc3cce43e96aad58a5868dc
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/68dec95a-37fb-4c8b-aa49-b265ecedcdf7/f9add4085095b9b6d28d045fd9c92c2c09f549a7
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/386a740c-7d17-4fc1-99f9-98919054ab98/12b23d207d23dd430b93320539abbb0bde84870d
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/fbf2bb40-979e-4ccb-9b6a-71d722c0fc56/4d5e7829ddc208866047fb6244c61dcb8d5db7aa
For , and denote the transpose and Hermitian transpose (also called conjugate transpose) respectively. If , then .
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/52190ec8-04ea-4095-8fdb-488b161a39d5/98b866c1c2d0c50c736f58199a1cdc7019d98f25
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/6326c08c-1039-4335-914e-8e69dbcb2262/0966b7ada93c12ff7d306496f224f7b3465a66d0
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/11e2fd20-ef8e-4552-8d82-dec2bdbfc956/44e23745a51c2c2d8d91fd98c1cf721573747ece
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/713ea0f2-6aef-4881-9ef7-086aabf47f28/aaa0cac96c9853f0cba61605679cf8963983b5d9
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/dd6a2ccc-ae5a-489f-a02c-d79d24687752/32df214ddc8971d6038cff8ef2db06cbf0b98b55
For , denotes the column space (image) of (the space spanned by the column vectors of ) and denotes the kernel (null space) of .
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/dbb7b370-6d17-47db-b6a3-68b9a9146be9/2661b1fb847ee5e1eb423afae01ad2288ee564d1
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/904f5d35-2389-46fe-847d-936a0b18ecf8/fab27fb4326a4df502f995954130806bc0311ecb
Finally, for any positive integer , denotes the identity matrix.
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/c31f3dda-1a09-43df-9cce-89d8aa3dcee4/a601995d55609f2d9f5e233e36fbe9ea26011b3b
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/85add689-1c70-4b79-95af-4a7d77ea1a6e/58cf11243afe4d043c27a89676d7f83f71310d82
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/5cb65df7-2cf7-4ea0-8f21-4dd50d5891c2/59d2b4cb72e304526cf5b5887147729ea259da78
For
, a pseudoinverse of
is defined as a matrix
https://s3-us-west-2.amazonaws.com/secure.notion-static.com/392d2978-001e-473b-bb3d-a785c90e72e1/aa629bc7b14c5bfc89aa06cfe4607cb560d71924
satisfying all of the following four criteria, known as the Moore–Penrose conditions:[7][8]
( need not be the general identity matrix, but it maps all column vectors of to themselves); ( acts like a [weak inverse](<https://en.wikipedia.org/wiki/Weak_inverse>)); ( is [Hermitian](<https://en.wikipedia.org/wiki/Hermitian_matrix>)); ( is also Hermitian).