Personal C++ implementations of machine learning primitives, built while working through the theory.

• perceptron/ — Single-layer perceptron with step activation.
• neuron/ — Single neuron with ReLU, tanh, and sigmoid activations.
• erm/ — Notes on Empirical Risk Minimization.

Parts of this codebase draw on:

• Shalev-Shwartz, S., & Ben-David, S. (2014). Understanding Machine Learning: From Theory to Algorithms. Cambridge University Press. Free PDF.

Some material follows this book; other parts come from other sources or my own exploration. See each subfolder's README for specifics. Credit for the referenced theory belongs to the authors.

See LICENSE.