← 🎛️ Machine Learning

🎛️ AutoML Search Lab

Trial searching
Evaluating config
Best model found
Trials evaluated: 0
Best accuracy:
Best learning rate:
Best model depth:
FPS:
Drag — rotate · Scroll — zoom

🎛️ AutoML Explained: How Machines Tune Machine Learning

Glowing search agents explore a 3D accuracy landscape shaped by two hyperparameters, hunting for the highest peak the way AutoML tools like random search, grid search and Bayesian optimisation hunt for the best model configuration.

🔬 What It Demonstrates

The terrain's height is validation accuracy across a learning-rate and model-depth grid, with gold marking the global optimum and lower peaks acting as misleading local optima. Each trial agent trains and evaluates a configuration, then reports its score before choosing the next one.

🎮 How to Use

Switch between random, grid and Bayesian-style search to compare coverage patterns, add more parallel trials, speed up the search, and raise landscape ruggedness to add more deceptive local optima. Watch the gold beacon jump each time a trial beats the current best.

💡 Did You Know?

Bergstra & Bengio's influential 2012 paper showed that plain random search often outperforms grid search of the same budget on high-dimensional problems — a key reason modern AutoML frameworks default to randomised or Bayesian strategies instead.