Backpropagation Training
Gradient-based algorithm for efficiently training multi-layer neural networks by propagating error derivatives backward.
Core metadata
- ID: backpropagation_training
- Era: Modern
- First known date: 1948 (decade)
- Region: Global / multiple regions
- Review status: source_checked
- Maturity: established
Prerequisites
- Algorithms & Computation Theory (algorithms_computation_theory)
- Logarithms & Early Calculus (logarithms_calculus_early)
- Perceptrons (perceptrons)
Dependents
Fields
Field lanes
- Artificial Intelligence & Machine Learning: Neural Networks
Node sources
- Learning Representations by Back-Propagating Errors (Nature, 1986, primary_paper) • Supports: node, maturity
Prerequisite edge evidence
Edge/source evidence summary:
- Prerequisite edges: 3
- Average edge confidence: 70%
- Prerequisite sources: 3
- expert_inference: 3
| Prerequisite | Type | Confidence | Evidence level | Note | Sources |
|---|---|---|---|---|---|
| Perceptrons (perceptrons) | enabling | 68% | expert_inference | Perceptrons provides a capability that enables this technology without being the only possible path. |
|
| Logarithms & Early Calculus (logarithms_calculus_early) | historical_predecessor | 75% | expert_inference | Logarithms & Early Calculus is an earlier historical predecessor or foundation, not a one-to-one engineering dependency. |
|
| Algorithms & Computation Theory (algorithms_computation_theory) | enabling | 68% | expert_inference | Algorithms & Computation Theory provides a capability that enables this technology without being the only possible path. |
|
This page is generated from canonical era JSON and is indexable by URL.