LESSON 1.5 · Code · 120 min
From Value to Neuron, Layer, and MLP
From Value to Neuron, Layer, and MLP. Learn parameter containers, module composition, and forward graphs and complete: Build an MLP in 50 lines. Part of the “Neural Network Foundat
Learning objectives
- Explain what problem “From Value to Neuron, Layer, and MLP” solves without hiding behind terminology.
- Trace the variables and causal links across parameter containers, module composition, and forward graphs.
- Complete “Build an MLP in 50 lines” and judge the result with evidence rather than intuition.
Core concepts
parameter containers
parameter containers 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.
module composition
module composition 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 forward graphs
and forward graphs 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 an MLP in 50 lines
- 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.