Across
- 3. Collection of data used to train machine learning models
- 5. When a model learns training data too well and fails on new data
- 6. Step-by-step procedure for solving a problem or performing a task
- 8. Systematic error in predictions or unfairness in model outcomes
- 9. Architecture revolutionizing NLP with attention mechanisms
- 13. Subset of machine learning using multi-layered neural networks
- 14. Using separate data to tune model parameters and prevent overfitting
Down
- 1. Task of categorizing data into predefined classes or groups
- 2. Process of teaching a model to recognize patterns in data
- 4. A computing system inspired by biological brain structure
- 7. Mathematical concept measuring rate of change in optimization
- 10. Prediction task involving continuous numerical output values
- 11. Learning approach where the model is trained with labeled examples
- 12. Multi-dimensional array used in modern ML frameworks