Across
- 3. A method in machine learning that allows models to be trained on local data from different devices without sharing sensitive data. (8)
- 4. Misleading or fabricated media generated through artificial intelligence techniques. (8)
- 6. The systematic error that occurs when an AI model's outcomes reflect human prejudices. (4)
- 7. The process of identifying patterns and features in inputs, often used in image processing. (10)
- 9. The most critical consideration in designing, developing, and implementing AI systems and models.(6)
- 10. A condition in which a machine learning model learns too much detail from training data and fails to generalize to new data. (12)
Down
- 1. The ability to describe how an AI system generates specific outputs or decisions. (13)
- 2. A term for the false outputs produced by generative models that can misrepresent reality. (12)
- 3. The practice of making predictions about future events based on historical data. (10)
- 5. The measure of impartiality in AI systems, ensuring equal treatment for individuals. (8)
- 8. A Networks encompasses algorithms that mimic the human brain's neuron connections. (14)