Neural-network building blocks · MLD-0652
Linear layer
A learned matrix transformation, usually with an optional additive bias.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Project a hidden vector into class scores.
Original caveat
With a bias, the mapping is affine rather than strictly linear.
Context and aliases
Level: Intermediate. Stages: Training · Inference · Theory.
Tags: DL · SLM · LLM · Vision · NLP.
Aliases: Dense layer · fully connected layer · affine layer.
Related concepts
Concepts linking here
Original references
- Deep Learning: online textbook ↗ · Consulted
- PyTorch neural-network layers ↗ · Reference link
- PyTorch documentation ↗ · Reference link
- Original reference URL ↗