Neural network output layers explained: classification vs regression
HarperSCarroll · x · 2026-11-09
AI educator Harper Carroll (Stanford) continues her neural network series explaining output layers:
- Two types of supervised models: classification models output probability per class (binary needs only one probability, the other is 1 minus it); regression models output a concrete number (e.g., house prices from zip code and square footage).
- Activation functions: the output layer squashes results to 0-1 as probability; hidden-layer activations are crucial for introducing non-linearity, since stacked weighted sums alone remain linear.
- Quiz: are LLMs classification or regression, and what do they predict?
A solid primer on the path from weighted sums to probability outputs.
More from Research
- Meta teases entry into AI-for-science race, joining frontier labs push — kimmonismus · 2026-11-09
- Study: 3g Daily Creatine for 4 Weeks Improved Cognition in Women, Most in Older Adults — 2C_ornot2C · 2026-11-09
- Brynjolfsson, Korinek, Agrawal publish Economics of Transformative AI research agenda in CACM — erikbryn · 2026-11-09
- Researchers question AI capability acceleration: shift to RL may be the confounder — DavidSKrueger · 2026-11-09
- Ropedia W0 trains robot world models on 10,000 hours of human video, lifting LIBERO-Plus by 6.76 points — jiqizhixin · 2026-11-09
- Mathematician rebuts AI proof critics: demanding Lean verification is a double standard — AlexKontorovich · 2026-11-09