Math
What is the degree of x^3y + 3xy^4 + z?
5
Explain the difference between "machine learning" and "artifical intelligence."
Machine learning is a subset of AI that involves systems that improve their performance without explicit instructions.
In which decade (+/- 1) was the ENIAC built?
1940s
What is the course number of "Introduction to Machine Learning" at Michigan?
445
Compute the following sum:
1 + 3 + 5 + ... + 99
2500
What is a hyperparameter?
The parameters in a model you set yourself, as opposed to the ones learned by the model.
In what year is Sophie graduating?
2026
Explain the difference between the validation and testing data sets.
Validation: used during development phase to assist in model selection and hyperparameter tuning
Testing: used after development to assess its performance on unseen data and measure its ability to generalize
What is the 18th derivative of cos(x)?
-cos(x)
What does i.i.d stand for?
Independent and identically distributed
How many years passed between the invention of the Transformer architecture and GPT?
1 year!
Attention Is All You Need (2017)
Improving Language Understanding by Generative Pre-Training (2018)
Which database of handwritten digits is commonly used in computer vision?
MNIST
State the first isomorphism theorem for vector spaces.
Let φ : V → W be a homomorphism between two vector spaces.
Then im φ is isomorphic to the quotient space V/ker(φ).
The Gaussian kernel is also called the RBF kernel. What does RBF stand for?
Radial basis function
Who was the principal investigator in the paper "Attention is All You Need?"
No one! Equal contribution from all authors.
What is the state-of-the-art scaling method for embedding weights in a click-through-rate prediction model?
η → η
λ → s^2λ