Lecture 1 (slides) Flashcards
1
Q
What is the general task of AI?
A
Predict output given specific features of the inputs
2
Q
How are AI and “regular” statsical modeling similair?
A
- Input features: independent variables
- Output class: dependent variable
- (in fact, ‘neural networks’ can be seen as a form of logistic regression models)
3
Q
What are the key differrences of AI to stastitical modeling?
A
- We care about predicting something, not about understanding a (causal) process
- Models are highly complex (and multicollinear) and generally seen as ‘black box’
4
Q
What is deep learning?
A
Fancy term for machine learning with very large models.
5
Q
On what is deep learning based on?
A
- Very large neural networks
- Trained on enormous amounts of data, e.g. “all of the internet”
- Using massive computing power, especially of GPUs
6
Q
What are key innovations for deep learning?
A
- Feature layers find patterns in raw input
- Networks can be (pre-)trained based on unannotated data
- Patterns from (pre-)training are transferred to actual task, and fine-tuned on annotated data
7
Q
What are the Core application area of AI?
A
Research field of NLP or Computational Linguistics
8
Q
Can the computer understand, generate, or translate text?
A
Generally
9
Q
What is a recent innovations that is based on ‘BERT’:
A
Decoder-encoder architecture that builds layers of understanding
trained using tasks such as ‘predict the next word’