Skip to content

CSCI3340

Scientific Computing

Floating-point arithmetic, direct and iterative linear solvers, least squares, eigenvalue computation, root finding, interpolation and quadrature, numerical ODEs and stochastic simulation.

10 chapters · 41 sectionsEN · 繁 · 简
My course notes

Your progress

0/ 41 sections marked complete

Next section: Floating point, rounding and numerical stability

Completion marks are saved in this browser.

This course in the web

In the CSCI3340 notes

Theorem cards
106
Definitions
84
Counterexamples
45
Concept lenses
10
Histories of ideas
1
Proof X-rays
43

Sections with a history of an idea

None of these is on the timeline yet.

No concept lens from this course is recorded in the web yet.

Open this course in the knowledge map

Counts cover the 41 public sections of this course and what is written in them, not recorded connections. 0 of these sections are indexed into the web in detail so far.

In the webTheorem cards 106 · Definitions 84 · Counterexamples 45 · Concept lenses 10 · Histories of ideas 1 · Proof X-rays 43

None of these is on the timeline yet.

No concept lens from this course is recorded in the web yet.

Open this course in the knowledge map

Counts cover the 41 public sections of this course and what is written in them, not recorded connections. 0 of these sections are indexed into the web in detail so far.

Chapter 1 · 4 sections

Floating-point arithmetic

Floating-point arithmetic

Theorems 11 · Definitions 11Counterexamples 4Proof X-rays 4

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 1.1

    Floating point, rounding and numerical stability

    Model rounding, analyze cancellation and distinguish conditioning from forward and backward error.

  2. 1.2

    Summation error and conditioning

    Separate the sensitivity of a sum from the rounding error of a particular accumulation order.

  3. 1.3

    Numerical differentiation and step size

    Balance truncation error against evaluation error instead of assuming a smaller step is always better.

  4. 1.4

    Compensated summation and its limits

    Trace a compensation term through rounded additions and distinguish arithmetic improvement from conditioning.

Chapter 2 · 5 sections

Linear systems

Linear systems

Theorems 9 · Definitions 7Counterexamples 6Lenses 1History 1Proof X-rays 5

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 2.1

    Linear systems: factorization, pivoting and conditioning

    Solve systems by elimination, interpret residuals, and distinguish stable algorithms from sensitive problems.

  2. 2.2

    Partial pivoting and element growth

    Prove the elementary growth bound for pivoted elimination and construct a sharp worst-case family.

  3. 2.3

    Rounding and backward error in pivoted LU factorization

    Track the arithmetic of elimination to connect computed factors, entrywise backward error and element growth.

  4. 2.4

    Cholesky factorization and Schur complements

    Prove the positive-definite factorization and see how it solves systems without forming an inverse.

  5. 2.5

    Normwise and componentwise backward error

    Find the smallest data perturbations that make an approximate linear-system solution exact, and relate them to conditioning.

Chapter 3 · 2 sections

Iterative solvers

Iterative solvers

Theorems 6 · Definitions 5Counterexamples 3Lenses 1Proof X-rays 2

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 3.1

    Jacobi, Gauss–Seidel, and successive over-relaxation

    Derive stationary iterations, prove convergence under explicit hypotheses, and compare rates and failure cases.

  2. 3.2

    Conjugate gradients and Krylov optimality

    Derive conjugate gradients for symmetric positive-definite systems, prove its energy minimization, and distinguish exact-arithmetic guarantees from computation.

Chapter 4 · 3 sections

Least squares

Least squares

Theorems 8 · Definitions 8Counterexamples 3Lenses 2Proof X-rays 3

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 4.1

    Least squares, QR factorization and numerical rank

    Derive projection and QR solutions, analyze sensitivity, and distinguish exact rank from numerical rank.

  2. 4.2

    Householder QR and stable reflections

    Construct QR with implicit orthogonal reflectors and understand the sign choice that avoids cancellation.

  3. 4.3

    Singular values, least squares and regularization

    Use the singular-value decomposition to derive minimum-norm least squares and compare truncation with ridge filtering.

Chapter 5 · 6 sections

Eigenvalue computation

Eigenvalue computation

Theorems 19 · Definitions 12Counterexamples 7Lenses 2Proof X-rays 6

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 5.1

    Gershgorin disks and eigenvalue bounds

    Locate eigenvalues using row sums and understand when a separated cluster has a known count.

  2. 5.2

    Power iteration and Rayleigh quotients

    Prove convergence under a dominant-eigenvalue hypothesis, examine failures, and interpret eigenpair residuals.

  3. 5.3

    Jacobi plane rotations for symmetric eigenproblems

    Construct a rotation, prove off-diagonal decrease for largest pivots, and interpret the resulting approximate eigenpairs.

  4. 5.4

    Inverse iteration and eigenvalue residuals

    Target an eigenvalue near a shift, and use the residual to certify proximity to the spectrum.

  5. 5.5

    QR iteration in two dimensions

    Connect repeated QR factorizations to simultaneous power iteration and see when eigenvalue convergence fails.

  6. 5.6

    Hessenberg reduction and shifted QR iteration

    Reduce a matrix to Hessenberg form by Householder reflectors, prove that shifted QR steps preserve it, connect the shift to inverse iteration, and use deflation to compute eigenvalues.

Chapter 6 · 3 sections

Root finding

Root finding

Theorems 9 · Definitions 7Counterexamples 4Lenses 1Proof X-rays 3

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 6.1

    Root finding: brackets, fixed points and Newton iteration

    Establish convergence conditions for bisection and Newton's method, and interpret stopping tests.

  2. 6.2

    The secant method and local superlinear convergence

    Derive a derivative-free root iteration and prove its local error product bound.

  3. 6.3

    Newton's method for nonlinear systems

    Derive the Newton step from a linear system and prove local quadratic convergence with explicit hypotheses.

Chapter 7 · 11 sections

Interpolation and quadrature

Interpolation and quadrature

Theorems 30 · Definitions 17Counterexamples 11Proof X-rays 11

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 7.1

    Interpolation and quadrature: from data to error bounds

    Construct interpolating polynomials, derive quadrature rules, and state what their error estimates require.

  2. 7.2

    Gaussian quadrature and orthogonal polynomials

    Derive optimal polynomial exactness from orthogonality and understand what it does and does not promise.

  3. 7.3

    Adaptive Simpson quadrature and error estimation

    Derive the difference-based estimator and distinguish an asymptotic indicator from a rigorous error certificate.

  4. 7.4

    Certified midpoint quadrature

    Turn curvature bounds and outward-rounded sample intervals into a provable enclosure of an integral.

  5. 7.5

    Cubic Hermite interpolation from values and slopes

    Construct an endpoint value-and-slope interpolant, prove its error bound, and track errors in supplied data.

  6. 7.6

    Newton divided-difference table

    Build, update and evaluate a Newton interpolant using a divided-difference table.

  7. 7.7

    Composite quadrature: signed error proofs

    Supplementary kernel proofs and solved practice; refinement examples overlap the existing adaptive Simpson note.

  8. 7.8

    Chebyshev nodes and minimax interpolation products

    Prove the extremal monic-polynomial property and use it to bound interpolation error.

  9. 7.9

    Barycentric interpolation and data conditioning

    Derive the barycentric formula, its node limits, and the exact sensitivity of interpolated values to data perturbations.

  10. 7.10

    Natural cubic splines

    Construct a twice continuously differentiable interpolant by solving for its nodal curvatures.

  11. 7.11

    Richardson extrapolation and Romberg integration

    Cancel a leading error term, derive the Romberg table, and identify when the cancellation is unreliable.

Chapter 8 · 5 sections

Numerical ODEs

Numerical ODEs

Theorems 9 · Definitions 13Counterexamples 5Lenses 1Proof X-rays 5

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 8.1

    Numerical ODEs: Euler, Runge–Kutta and stability

    Derive first- and second-order methods, prove convergence under explicit assumptions, and analyze stability on a test equation.

  2. 8.2

    Two-step Adams–Bashforth: accuracy and stability

    Derive a multistep method and separate its local error, global convergence and absolute stability.

  3. 8.3

    Implicit Euler, trapezoidal steps and stability

    Derive two implicit integrators, prove their stability properties and distinguish damping from accuracy.

  4. 8.4

    Order conditions for linear multistep methods

    Derive the coefficient moment conditions, calculate local defects, and separate consistency from zero-stability.

  5. 8.5

    Multistep root conditions and zero-stability

    Understand how parasitic roots amplify starting and truncation errors in linear multistep methods.

Chapter 9 · 1 section

Stochastic simulation

Stochastic simulation

Theorems 2 · Definitions 3Counterexamples 1Lenses 1Proof X-rays 2

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 9.1

    Monte Carlo estimates and importance sampling

    Prove convergence and finite-sample bounds for independent simulations, and analyze the variance and support requirements of importance sampling.

Chapter 10 · 1 section

Sparse stationary iterations

Iteration dependencies, convergence and error certificates.

Theorems 3 · Definitions 1Counterexamples 1Lenses 1Proof X-rays 2

Chapter outline →See this chapter on the web →Not yet indexed in the web

  1. 10.1

    Jacobi and Gauss–Seidel: convergence and certified stopping

    Track sweep dependencies and derive step-based and residual-based error guarantees.

Vercel visit statistics and public-page performance measurements each require separate consent. Neither is needed to use the website.

Website function storage (used when needed)

Security, sign-in and your chosen preferences use storage when needed. This is separate from optional statistics.

Withdrawing stops future statistics. The page reloads if analytics code needs unloading. It does not delete data already sent.

Read Cookie and storage details