Lecture Notes on Artificial Intelligence

These are the resources for my lecture on artificial intelligence. The directory Lecture-Notes contains the LaTeX files, while the directory Python contains jupyter notebooks that implement the algorithms discussed in my lecture.

Download the lecture notes: PDF or EPUB. Both finished books are included in the repository; no build tools are needed to read them.

Export PDF and EPUB

macOS setup

Set up Homebrew, including its shell instructions, then install full MacTeX and the export tools:

xcode-select --install                 # if Command Line Tools are missing
brew install --cask mactex-no-gui       # skip if current full MacTeX is installed
brew install python ghostscript mupdf-tools librsvg epubcheck
export PATH="/Library/TeX/texbin:$(brew --prefix)/bin:$PATH"

Use Python 3.10+ and EPUBCheck 5.4+; Homebrew's EPUBCheck package includes Java. On other systems, use full TeX Live plus Python, Ghostscript, MuPDF (mutool), librsvg (rsvg-convert), ZIP and EPUBCheck with Java. All tools must be on PATH; no Python packages are required.

Build

From the repository root:

make -C Lecture-Notes

This builds Lecture-Notes/artificial-intelligence.pdf and Lecture-Notes/artificial-intelligence.epub from the same LaTeX source. Append pdf or epub to build one format, check-epub to validate an existing EPUB, or clean to remove temporary files.

If a command is missing, check the installations and PATH above. EPUB build diagnostics are in Lecture-Notes/.build/epub/export.log; a failed export preserves the previous EPUB. Nonstandard Ghostscript installations may need LIBGS.

The EPUB is reflowable and works offline, with searchable text, navigation, code, linked figures/equations, bibliography, footnotes and index. Mathematics and diagrams use SVG with text alternatives; the PDF retains its print layout. EPUBCheck must pass without warnings before publication. SVG is not structured MathML, and reader support varies; check the book in the intended reading applications before release. PDF figures are converted with MuPDF to retain embedded bitmap artwork. The cover is a PNG rendered from Lecture-Notes/epub/cover.svg for reader compatibility.