Across
- 3. One complete pass through the entire training dataset
- 4. The ability of a model to perform well on new, unseen data
- 6. A step-by-step procedure that computers follow to solve problems
- 8. When a model performs too well on training data but poorly on new data
- 9. A systematic error that occurs in machine learning predictions
Down
- 1. Basic processing unit in artificial neural networks, inspired by brain cells
- 2. The process of teaching a model using historical data
- 3. A measurement of how wrong a model's predictions are
- 5. The process of splitting input data into smaller subsets for training
- 7. A mathematical representation that makes predictions based on input data
