Across
- 1. Instructions or input given to an AI model to generate a response
- 5. A type of neural network that compresses data into a smaller representation and then reconstructs it
- 7. The process of converting text or data into numerical form or so a machine can process it
- 8. A complete pass of the training dataset through machine learning model
- 9. Using technology to perform tasks without human input
- 10. The process of using a trained model to make predictions on new data
- 11. The ability of a model to perform well on unseen data, not just the training set
- 15. When a model learns noise in the data instead of the actual pattern
- 16. A set-by-step set of instructions for solving a problem
- 18. A dense vector representation of words, images, or data features in a lower-dimensional space
- 19. Settings adjusted to optimize AI models
- 20. Field of AI that processes and understands human language
Down
- 2. AI that improves by using examples without being told how
- 3. Systematic error in AI predictions due to skewed data
- 4. Branch of AI that deals with intelligent machines
- 6. Breaking text into smaller units for analysis
- 12. Principles ensuring responsible and fair AI usage
- 13. AI program designed to simulate human conversation
- 14. A technique to reduce model complexity and prevent overfitting
- 17. Model inspired by the human brain for pattern recognition