GW Gregorius Willson 05 / 06 · Project
Interactive ML
Browser laboratory for machine-learning intuition
Problem
Loss functions, optimizers, clustering, and decision boundaries are difficult to internalize from static explanations alone.
Approach
Built a client-side laboratory where parameters can be changed directly and the resulting model behavior, boundaries, and optimization paths update immediately.
Result
A public interactive learning site that makes core machine-learning mechanics visible without requiring a backend or local setup.
Live
Live parameter tuning
Client side simulation
A fully client-side laboratory where model parameters, optimization, and decision boundaries respond in real time.