Concept dictionary
Neural Network
A neural network is a model made of connected layers of numerical operations whose adjustable weights are learned from data.
- Review state
- Reviewed in POC
- Freshness
- Current
- Version
- 2
Also called
- Neural Network
- Artificial Neural Network
Remember it this wayConnected numerical layers learn how strongly signals should matter.
In plain language
Information passes through connected units that transform it. Training adjusts the strength of those connections so the network becomes better at a defined task.
Why it matters
Neural networks are a core building block of deep learning and many current AI systems. Their performance depends on architecture, data, training choices, and evaluation—not the name alone.
Example
A neural network receives measurements from a machine and estimates whether the equipment is operating normally or showing signs of failure.
Common misconception
The word neural does not mean the system contains biological neurons or thinks like a human brain. It is a mathematical model with limited biological inspiration.
Related terms
Sources
- Google for Developers, Machine Learning Glossarytechnical definition and examples of neural networks.
- NIST AI 100-2 E2025, Adversarial Machine LearningNIST terminology for neural-network models.
Learn more
- Follow the learned values through [[weights|Weights]] and [[parameters|Parameters]].