Applications of Vector Spaces and Subspaces
Monserrath Trujillo, Angely Aguilar, Esteban Ortiz, Andy
Vargas, Josue Marquina
May 30
Image Processing and Compression - JPEG
Discrete Cosine Transform (DCT) in JPEG
▶ Concept:
▶ The DCT converts a signal (such as an image) from the
spatial domain to the frequency domain.
▶ Application:
▶ In JPEG compression, the image is divided into 8x8 pixel
blocks.
▶ Each block is transformed using the DCT.
▶ The resulting coefficients are quantized, reducing or
eliminating high-frequency coefficients.
▶ The compressed image is reconstructed using the inverse DCT.
▶ Problem Solving:
▶ Reduces storage size and bandwidth required to transmit
images without significant loss of visual quality.
Algorithms and Computing - Singular Value Decomposition
(SVD)
Concept:
▶ SVD decomposes a matrix into three matrices: a diagonal and
two orthogonal ones.
Application:
▶ In movie recommendations (e.g., Netflix), user rating matrices
are decomposed to identify latent patterns and relationships.
Problem Solving:
▶ Recommends movies to users based on their preferences and
those of similar users, enhancing the user experience.
Economics and Finance - Portfolio Optimization
Concept:
▶ Uses Markowitz’s portfolio theory, which is based on vector
spaces to represent and manage the combination of assets.
Application:
▶ An investor can define a set of financial assets and their
expected returns, covariances, and variances.
Problem Solving:
▶ Constructs a portfolio that maximizes expected return for a
given level of risk or minimizes risk for a given level of return,
aiding informed investment decisions.
Computer Graphics - Geometric Transformations
Concept:
▶ Transformations such as rotations, translations, and scalings
are represented by matrices.
Application:
▶ In 3D rendering, each point of an object is multiplied by
transformation matrices to change its position and orientation
in space.
Problem Solving:
▶ Manipulates and visualizes 3D objects in computer-aided
design (CAD) applications and video games.
Physics and Quantum Chemistry
Quantum Mechanics
▶ Concept:
▶ Quantum states are represented as vectors in a Hilbert space,
and quantum operators are linear operators in these spaces.
▶ Application:
▶ Used to describe quantum systems, such as electrons in an
atom, using wave functions that are solutions to the
Schrödinger equation.
▶ Problem Solving:
▶ Predicts and explains the behavior of subatomic particles and
develops technologies like nuclear magnetic resonance (NMR)
and quantum computers.
Spectroscopy
Concept:
▶ The interaction of light with matter is modeled using linear
operators in Hilbert spaces, analyzing transitions between
energy levels.
Application:
▶ Used to identify and quantify substances based on their
absorption or emission spectra.
Problem Solving:
▶ Helps determine the chemical composition of samples, study
molecular structures, and diagnose diseases using techniques
such as infrared (IR) spectroscopy and magnetic resonance
(MR) spectroscopy.
Applications of Vector Spaces Using Parameterization
Computer Graphics and 3D Modeling
▶ The parameterization of surfaces and trajectories plays a
crucial role:
▶ Surface Modeling: Allows representing 3D objects like the
cylinder parameterized by r (u, v ) = (r cos u, r sin u, v ),
facilitating the creation of realistic graphic models.
▶ Animation: Object trajectories are parameterized using
r (t) = r0 + tv , where r is the initial position and v the
velocity, enabling smooth and precise movements in animation
and simulation applications.
These techniques are fundamental in the entertainment industry,
computer-aided design (CAD), and virtual simulations, providing
powerful tools for creating and manipulating objects in 3D
environments.
Applications of Kernel, Domain, and Range in Vector
Spaces
Signal Processing
▶ Concept:
▶ Use of linear transformations to analyze signals.
▶ Domain:
▶ All input signals that can be processed.
▶ Kernel:
▶ Signals that result in a null signal.
▶ Range:
▶ All possible output signals.
▶ Importance:
▶ Essential for filter design and signal analysis.
Differential Equations
Concept:
▶ Vector spaces describe the solutions of differential equations.
Domain:
▶ All initial conditions and time values.
Kernel:
▶ Homogeneous solutions that nullify.
Range:
▶ Particular solutions of the system.
Importance:
▶ Fundamental for modeling and simulation of physical systems.
Conclusion
Summary:
▶ Vector spaces and their properties, such as domain, kernel,
and range, are fundamental in various practical applications.
▶ These concepts allow breaking down complex problems into
more manageable and understandable components.
▶ We have seen how these concepts are applied in areas as
diverse as image processing (improving data compression and
transmission), optimal controller design (enhancing efficiency
and precision in system control), recommendation algorithms
(improving user experience), and financial optimization
(aiding informed investment decisions).
