# Ai Foundations Cheat Sheet

| Term | Practical meaning |
|---|---|
| Artificial Intelligence | Systems performing tasks involving perception, prediction, reasoning, generation, decisions or action. |
| Algorithm | A defined method or procedure for solving a problem. |
| Model | A learned or designed representation used to produce an output. |
| Machine Learning | Methods that learn patterns from data. |
| Deep Learning | Machine Learning using neural networks with multiple processing layers. |
| Generative Ai | Systems that generate new text, images, audio, video or other content. |
| Training | Adjusting a model using data or experience. |
| Validation | Comparing development choices without using the final test evidence. |
| Testing | Final independent evaluation on unseen examples. |
| Inference | Using a trained model to produce an output. |
| Feature | An input variable or representation used by a model. |
| Label | A known target or category used in supervised learning. |
| Parameter | A value learned by the model. |
| Hyperparameter | A setting chosen before or during training. |
| Overfitting | Learning training-specific detail that does not generalize. |
| Generalization | Performing usefully on new, relevant data. |

## Important distinctions

- Algorithm ≠ model: the method learns or constructs the model.
- Ai ≠ automation: automation can follow fixed rules without learning.
- Accuracy ≠ usefulness: a technically strong result may still fail users.
- Generated fluency ≠ factual correctness: verify important claims.