LESSON 1.1 · Intuition · 75 min
A neuron is not magic: functions, parameters, activations
A neuron is not magic: functions, parameters, activations. Learn linear combinations, nonlinearity, expressiveness, and decision boundaries and complete: Build a two-input neuron b
Learning objectives
- Explain what problem “A neuron is not magic: functions, parameters, activations” solves without hiding behind terminology.
- Trace the variables and causal links across linear combinations, nonlinearity, expressiveness.
- Complete “Build a two-input neuron by hand” and judge the result with evidence rather than intuition.
Core concepts
linear combinations
linear combinations is part of the lesson’s causal model. State its inputs, outputs, invariants, and failure mode; then verify it with a hand-check or a minimal experiment before moving to an optimized implementation.
nonlinearity
nonlinearity is part of the lesson’s causal model. State its inputs, outputs, invariants, and failure mode; then verify it with a hand-check or a minimal experiment.
expressiveness
expressiveness is part of the lesson’s causal model. State its inputs, outputs, invariants, and failure mode; then verify it with a hand-check or a minimal experiment.
and decision boundaries
and decision boundaries is part of the lesson’s causal model. State its inputs, outputs, invariants, and failure mode; then verify it with a hand-check or a minimal experiment.
Build and verify
Build a two-input neuron by hand
- Predict: write the expected output, trend, or failure before running code.
- Build: implement only the minimum components needed to answer the question.
- Verify: compare with a baseline or trusted implementation; save seeds, parameters, and raw outputs.
- Transfer: change one shape, dataset, scale, or workload condition and explain whether the conclusion still holds.