Scribd
Upload
4 views9 pages

Guru

The document discusses orthogonal projections and the Gram-Schmidt process in linear algebra, highlighting their definitions, properties, and applications. Orthogonal projections are used for decomposing vectors and solving least squares problems, while the Gram-Schmidt process is essential for creating orthogonal bases and QR factorization. Both concepts are crucial in various fields such as data fitting, computer graphics, signal processing, machine learning, and quantum mechanics.

Uploaded by

purohitsudhanva05
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
4 views9 pages

Guru

The document discusses orthogonal projections and the Gram-Schmidt process in linear algebra, highlighting their definitions, properties, and applications. Orthogonal projections are used for decomposing vectors and solving least squares problems, while the Gram-Schmidt process is essential for creating orthogonal bases and QR factorization. Both concepts are crucial in various fields such as data fitting, computer graphics, signal processing, machine learning, and quantum mechanics.

Uploaded by

purohitsudhanva05
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Dayananda Sagar college of Engineering

(An Autonomous Institute Affiliated to VTU, Belagavi, Accredited by NAAC with ‘A’ Grade)
Shavige Malleshwara Hills, Kumaraswamy Layout, Bengaluru-560111

Topic Report
Linear Algebra – Orthogonal Projections and Gram-Schmidt Process Submitted by
Name USN
Sudhanva S P 1DS23EE105
T Veerabala 1DS23EE119
Hemanth 1DS23EE406
Pradyumna N 1DS23EE408

4Th Sem in Electrical and Electronics Engineering


Course Name : Probability Theory and Linear Algebra (22MAT41B)
Under the Guidance of
Prof. Komala C S
Assistant Professor
Department of Mathematics
DSCE , Banglore
 Introduction to Orthogonal Projections and the Gram-Schmidt Process

In linear algebra, orthogonal projections and the Gram-Schmidt process are fundamental concepts used to work with
orthogonal and Orthonormal in vector spaces.

1. Orthogonal Projection: An orthogonal projection maps a vector onto a subspace in such a way that the difference
between the original vector and its projection is orthogonal to the subspace. This is useful for:

• Decomposing vectors into parallel and perpendicular components.


• Solving least squares problems.
• Applications in computer graphics, signal processing, and machine learning.
2. Gram Schmidt Process: It is an algorithm that converts a set of linearly independent vectors into an orthogonal basis
for their span.

This process is essential for:

• Constructing orthogonal bases for subspaces.


• QR factorization of matrices.
• Numerical stability in computations (e.g., in eigenvalue algorithms).
 Properties of Orthogonal Projections and the Gram-Schmidt Process

• Properties of Orthogonal Projections


1. Linearity: The orthogonal projection operator PP is a linear transformation, meaning:
2. P(au+bv)=aP(u)+bP(v)P(au+bv)=aP(u)+bP(v)
3. Idempotence: Applying the projection twice does not change the result:
4. P2=PP2=P
5. Orthogonality Condition: The residual vector v−P(v)v−P(v) is orthogonal to the subspace onto which (
\mathbf{v} is projected.
6. Range and Null Space:
7. The range of PP is the subspace onto which we project.
8. The null space of PP is the orthogonal complement of this subspace.
9. Symmetry (for real spaces): If PP is a projection onto a subspace WW, then PP is symmetric
(i.e., PT=PPT=P) with respect to the standard inner product.
• Properties of the Gram-Schmidt Process

1. Preserves Span: The orthogonal (or orthonormal) vectors produced by Gram-Schmidt span the same
subspace as the original vectors.
2. Uniqueness (up to scaling): The orthogonal basis obtained is unique if we enforce normalization, but
different normalizations lead to different orthonormal bases.
3. Numerical Instability: The classical Gram-Schmidt process can suffer from loss of orthogonality due to
rounding errors in floating-point arithmetic. Modified versions (e.g., Modified Gram-Schmidt) improve
stability.
4. QR Factorization Connection: Gram-Schmidt is related to the QR decomposition of a matrix,
where A=QRA=QR (QQ is orthogonal, RR is upper triangular).
5. Orthonormal Output: If the process is carried out with normalization, the resulting basis vectors are
orthonormal (unit length and mutually orthogonal).
 Applications of Orthogonal Projections
Orthogonal projections involve projecting a vector onto a subspace such that the projection is orthogonal (perpendicular) to
the subspace. Applications include:

a. Least Squares Approximation (Data Fitting)

• Used in regression analysis (linear, polynomial) to minimize the error between observed data and predicted values.
• Helps solve overdetermined systems Ax=bAx=b by projecting bb onto the column space of AA.

b. Computer Graphics & 3D Rendering

• Projecting 3D objects onto 2D screens while preserving perspective.


• Shadow generation by projecting objects onto surfaces.

c. Signal Processing

• Noise reduction by projecting a noisy signal onto a subspace of desired signals.


• Used in Fourier analysis to decompose signals into orthogonal components.
d. Machine Learning & Data Science

• Principal Component Analysis (PCA) uses projections to reduce dimensionality while preserving variance.
• Feature extraction by projecting data onto meaningful subspaces.

e. Control Theory

• State estimation and filtering (e.g., Kalman filters use projections to estimate system states).

 Applications of Gram-Schmidt process


a. QR Decomposition (Numerical Linear Algebra)

• Factorizing a matrix AA into QRQR, where QQ is orthogonal and RR is upper triangular.


• Used in solving linear systems, eigenvalue computations, and matrix inversion.

b. Orthogonal Polynomials

• Constructing orthogonal polynomial sets (Legendre, Chebyshev, Hermite) used in:


• Numerical integration (Gaussian quadrature).
• Solving differential equations (Sturm-Liouville problems).
c. Computer Vision & Robotics

• Orthonormalizing basis vectors in pose estimation and camera calibration.


• Ensuring numerical stability in transformations.

d. Quantum Mechanics

• Constructing orthonormal bases for quantum states (wavefunctions).


• Used in the diagonalization of Hamiltonians.

e. Statistics & Econometrics

• Ensuring independence in regression variables.


• Orthogonalizing predictors in multivariate analysis.
Problem on the Gram-Schmidt process

Application: Orthonormalizing Quantum State ∣ψ2′⟩=∣ψ2⟩−(⟨ψ1∣ψ2⟩)∣ψ1⟩=[1212]−(12)[10]=[012]∣ψ2′​⟩=


Preparation ∣ψ2​⟩−(⟨ψ1​∣ψ2​⟩)∣ψ1​⟩=[2​1​2​1​]−(2​1​)[10​]=[02​1​]Step
3: Normalize ∣ψ2′⟩∣ψ2′​⟩ to get:
• Problem: Quantum algorithms (like Grover's or
Shor's) require well-defined orthogonal states. ∣ψ2orth⟩=[01]∣ψ2orth⟩=[01​]Now, {∣ψ1⟩,∣ψ2orth⟩}{∣ψ1​⟩,∣
However, due to noise or imperfect gates, quantum ψ2orth⟩} forms an orthonormal basis, ensuring
registers may end up in non-orthogonal states. reliable quantum operations.
• Solution: Apply the Gram-Schmidt process to
Why This Matters in Quantum Computing:
enforce orthogonality before further operations.
• Prevents errors in quantum gates (e.g., when
Example: Correcting a 2-Qubit State implementing controlled operations).
Suppose we have two (non-orthogonal) quantum states: • Used in quantum error correction to maintain state
purity.
∣ψ1⟩=[10],∣ψ2⟩=[1212]∣ψ1​⟩=[10​],∣ψ2​⟩=[2​1​2​1​]Step
1: Normalize ∣ψ1⟩∣ψ1​⟩ (already normalized). • Essential for quantum tomography (reconstructing
Step 2: Remove the projection unknown states).
of ∣ψ2⟩∣ψ2​⟩ onto ∣ψ1⟩∣ψ1​⟩:
Conclusion

 Orthogonal projections and the Gram-Schmidt process are essential in linear algebra, with applications in data
fitting (least squares), computer graphics, signal processing, machine learning (PCA), and quantum mechanics.
These tools simplify computations, enhance numerical stability, and enable efficient solutions across science and
engineering. Their versatility makes them fundamental in both theory and real-world problem-solving.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy