Compute the dot product of
<2x - 2, 3y>
<y + 6, x + 3>
5xy + 12x + 7y - 12
In 2018, which prestigious prize was awarded to three researchers for their work in deep learning?
Turing Award
Which model developed by OpenAI is designed for complex reasoning tasks?
o1
Who is the president of Michigan Hackers?
Matthew Zepf
Does the following series converge?
1 + 1/4 + 1/9 + 1/16 + ...
(For +100: If not, explain why. If so, to what value?)
Yes, to pi/6.
Who won the 2024 Nobel Prize in Physics?
Hopfield and Hinton
What is the difference between fine-tuning and distilling a model?
Distillation reduces the number of parameters to create a smaller model - fine-tuning adjusts existing parameters for a specific task.
Vector-quantized generative adversarial network
Shuffle a standard deck of cards with two jokers. What is the expected number of cards between the jokers?
52/3
In what year was PyTorch released?
2016
What does CLIP stand for?
Contrastive language-image pretraining
What is the difference between L1 and L2 regularization?
L1 penalizes based on the absolute value of the model coefficients, while L2 penalizes based on the square of the coefficients.
State the structure theorem for finitely generated abelian groups.
A group is a finitely generated abelian group if and only if if is isomorphic to
Z_{n1} * ... * Z_{nk} * Z^r
for some list of powers of primes n1, ..., nk.
In what year was Claude Shannon born? (+/- 5)
1916
What kind of positional encoding is used in the original Transformer paper?
Sinusoidal
Name one of the F23 MHML leads.
Omkar Yadav, Brandon Xu