Optimization and training dynamics · MLD-1171
Stochastic gradient descent
Gradient-based updates using a randomly sampled example or subset rather than the full dataset.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Train from shuffled mini-batches.
Original caveat
Common implementations use mini-batches rather than single examples.
Context and aliases
Level: Intermediate. Stages: Training · Theory.
Tags: ML · DL · SLM · LLM.
Aliases: SGD.
Related concepts
Concepts linking here
Original references
- Deep Learning: online textbook ↗ · Consulted
- PyTorch optimization algorithms ↗ · Reference link
- PyTorch documentation ↗ · Reference link
- Original reference URL ↗