Across
- 3. Type of understanding where models consider the context of words in sentences.
- 7. An instruction given to AI Model.
- 10. Optimization algorithm to minimize errors.
- 11. Training a model on unlabeled data.
- 12. Learning method where an agent learns by interacting with its environment.
- 15. Large and complex data sets.
- 17. Parameter within a neural network adjusted during training.
- 18. Computing system inspired by the human brain's network of neurons.
- 19. A step-by-step procedure for solving a problem or accomplishing a task
- 20. Model performs well on training data but poorly on unseen data.
- 21. Algorithms that allow computers to learn from data.
Down
- 1. Training a model on labeled data.
- 2. AI program that simulates human conversation.
- 4. One complete pass through the training dataset.
- 5. The multi-dimensional space where data is mapped in machine learning.
- 6. An individual measurable property of data.
- 8. A collection of data used for training models.
- 9. Subset of machine learning using neural networks with many layers.
- 13. A method to measure the model's error.
- 14. Data used to tune model parameters.
- 16. Settings that control the model training process.