GS
Grace Sullivan
last week
Practical and rigorous at the same time. I went straight from reading a capsule to shipping it at work.
Free preview available. Sign in and subscribe to unlock the full book.
Vizuara AI Labs · advanced
Scientific Machine Learning (SciML)
Physics-informed neural networks, Neural ODEs, and UDEs in Julia.
Learn Scientific Machine Learning from the math up: the three pillars — PINNs, Neural ODEs, and Universal Differential Equations — built by hand in Julia. Then take a real project all the way to a publishable first-author research paper.
39 capsules141 figures~8 hoursby Dr. Raj Dandekar
Read on your Kindle or e-reader
Download the EPUB and read offline — perfect for the train. Works on Kindle, Kobo, Apple Books & more.
00Foundations of Scientific ML5 capsules
Foundations of Scientific ML — 5 chapters.
01Why Scientific Machine Learning?intuitionfree12 min02Transitioning into ML from a Science Backgroundconceptfree11 min03The SciML Landscape: A Technical Overviewconcept🔒12 min04ODEs and PDEs in One Pagemath🔒12 min05The Problems SciML Can Solveintuition🔒14 min01The Julia Language for SciML4 capsules
The Julia Language for SciML — 4 chapters.
06Why Julia for Scientific MLconcept🔒10 min07Installing Julia and Meeting the REPLcode🔒11 min08Project Environments and the Package Managercode🔒10 min09Julia Fundamentals for Numericscode🔒13 min02Differential Equations in Julia6 capsules
Differential Equations in Julia — 6 chapters.
10Solving Your First ODE in Juliacode🔒13 min11Systems of ODEs and Choosing a Solvercode🔒11 min12PDE Mathematics and Boundary Conditionsmath🔒12 min13The Method of Linesmath🔒13 min14Solving the Heat and Brusselator PDEscode🔒13 min15Project: Solving the Schrodinger Equationproject🔒13 min03Neural Network Foundations4 capsules
Neural Network Foundations — 4 chapters.
16Feedforward Neural Networksconcept🔒11 min17Gradient Descent: A Mountain Perspectiveintuition🔒12 min18Backpropagation from Scratchmath🔒11 min19Automatic Differentiation: The Engine of SciMLdeep-dive🔒13 min04Pillar 1 — Physics-Informed Neural Networks4 capsules
Pillar 1 — Physics-Informed Neural Networks — 4 chapters.
20PINNs: Encoding Physics in the Lossconcept🔒10 min21The Math of the PINN Residualmath🔒12 min22Building Your First PINN in Juliacode🔒13 min23Project: The 1D Wave Equation with a PINNproject🔒15 min05Pillar 2 — Neural ODEs4 capsules
Pillar 2 — Neural ODEs — 4 chapters.
24Neural ODEs: Networks as Continuous Dynamicsconcept🔒12 min25Training Neural ODEs: The Adjoint Methodmath🔒11 min26Building Your First Neural ODE in Juliacode🔒13 min27Project: Neural ODEs and Their Applicationsproject🔒11 min06Pillar 3 — Universal Differential Equations4 capsules
Pillar 3 — Universal Differential Equations — 4 chapters.
28UDEs: Known Physics Plus a Learned Termconcept🔒10 min29From Learned Term to Discovered Equationdeep-dive🔒10 min30Building Your First UDE in Juliacode🔒13 min31Project: Recovering Predator-Prey Dynamicsproject🔒12 min07Hands-On Research Projects4 capsules
Hands-On Research Projects — 4 chapters.
32Framing a Scientific Problem as a SciML Problemintuition🔒12 min33Research Project: Covid-19 Prediction with a Quarantine-SIR UDEproject🔒10 min34Research Project: Black Hole Dynamics with SciMLproject🔒12 min35Reading and Reusing the Project Codecode🔒11 min08From Results to Published Research4 capsules
From Results to Published Research — 4 chapters.
36From a Project to a Research Paperconcept🔒11 min37Crafting Figures and the Scientific Narrativeintuition🔒11 min38Choosing a SciML Research Topicconcept🔒11 min39The SciML Capstone: Putting It All Togetherproject🔒12 minRatings & reviews
743 readersRate this book
AI
Ananya Iyer
last month
This is the resource I wish I had when I started. Clear mental models, zero fluff.
JB
Jonas Bergström
2 months ago
Beautifully produced and genuinely deep. The reader experience makes it easy to keep going for hours.