Introduction to Ordinary Differential Equations

2025-fall, Università della Svizzera italiana, Faculty of Informatics, 2025

This course introduces the fundamental concepts and techniques of ordinary differential equations (ODEs), with a focus on applications in artificial intelligence and computational science. We connect modeling and analysis (existence, stability, qualitative behavior) with computation (numerical initial-value solvers, stiffness) and modern ODE-based learning (Neural ODEs, sparse discovery, and multiscale/network dynamics).

For up-to-date information on schedule and lecture room, visit the official USI course page.

Schedule (14 weeks)

  • Format: 1 lecture per week on Fridays. After every two lectures there is an Exercises + Quiz session (first half TA Q&A, second half quiz). After Quiz 4 there is one last lecture and a final Exercises session before the exam.
WeekDateTopicMaterialsNotes
12025-09-19Lecture 1: Introduction and First-Order EquationsNotes 
22025-09-26Lecture 2: Systems of First-Order EquationsNotes 
32025-10-03Exercises + Quiz 1 TA Q&A + Quiz
42025-10-10Lecture 3: Linear Systems and Matrix MethodsNotes 
52025-10-17Lecture 4: Eigenvalue Methods and DiagonalizationNotes 
62025-10-24Exercises + Quiz 2 TA Q&A + Quiz
72025-10-31Lecture 5: Nonlinear Dynamics and Phase Plane AnalysisNotes 
82025-11-07Lecture 6: Stability Theory and Lyapunov MethodsNotes 
92025-11-14Exercises + Quiz 3 TA Q&A + Quiz
102025-11-21Lecture 7: Numerical Methods for Differential EquationsNotes 
112025-11-28Lecture 8: Applications in Science and EngineeringNotes 
122025-12-05Exercises + Quiz 4 TA Q&A + Quiz
132025-12-12Lecture 9: Advanced Topics and Current ResearchNotes 
142025-12-19Final Exercises & Q&A (exam preparation) No quiz

Slight adjustments may occur; any changes will be announced here and in class.

Course Outline (Lectures 1–9)

  1. Lecture 1 — Introduction and First-Order Equations: Basic concepts, classification of ODEs, direction fields, isoclines, separable equations, and linear first-order equations with integrating factors.
  2. Lecture 2 — Systems of First-Order Equations: Reduction of higher-order equations, phase space, trajectories, nullclines, and linearization for analyzing equilibria in systems like predator-prey models.
  3. Lecture 3 — Linear Systems and Matrix Methods: The matrix exponential, eigenvalue analysis for classifying 2D systems (nodes, saddles, spirals), fundamental matrices, and nonhomogeneous systems.
  4. Lecture 4 — Eigenvalue Methods and Diagonalization: Solving systems via diagonalization and modal coordinates, handling complex and repeated eigenvalues, and applications to mechanical systems.
  5. Lecture 5 — Nonlinear Dynamics and Phase Plane Analysis: In-depth analysis of nonlinear systems, limit cycles, Poincaré-Bendixson theorem, and local bifurcations (saddle-node, transcritical, pitchfork, Hopf).
  6. Lecture 6 — Stability Theory and Lyapunov Methods: Rigorous definitions of stability, Lyapunov’s direct method, LaSalle’s invariance principle, estimating basins of attraction, and stability of periodic orbits.
  7. Lecture 7 — Numerical Methods for Differential Equations: Euler’s method, Runge-Kutta methods, multi-step methods, local vs. global error, stability regions, stiffness, adaptive step-size control, and geometric integration.
  8. Lecture 8 — Applications in Science and Engineering: Broad survey of ODE modeling in mechanics, population dynamics (logistic growth, SIR models), biochemical networks, and coupled oscillator systems.
  9. Lecture 9 — Advanced Topics and Current Research: Modern developments including Neural ODEs with adjoint sensitivity, sparse discovery of dynamics (SINDy), consensus on graphs, multiscale methods, and continuous normalizing flows.