Computing · AI foundations / Principles and exploration
Perceptron classification experiment guide
Vary learning rate and data shape to watch the decision boundary move.
How does it work?
Linearly separable data can be learned with a straight boundary.
Key relationship
w ← w + η(y − ŷ)x · b ← b + η(y − ŷ)
Try this investigation
- Vary learning rate and data shape to watch the decision boundary move.
- Run and record each of the three reference settings.
- Change one variable and compare with the reference records.
Interpreting your results
At 100% accuracy, how many errors remain?
Zero · Linearly separable data can be learned with a straight boundary.
Assumptions and limitations
This is a single-layer binary classifier. XOR is not linearly separable. Initialization and order affect convergence.