Review
What is overfitting?
Overfitting occurs when a model learns too much from its training data, making it unable to generalize to new data.
What does LLM stand for?
Large language model
Compute d/dx sin(x)cos(x).
cos(x)^2 - sin(x)^2
Who are we playing against in March Madness tonight?
UCSD @ 10PM
What does the gradient represent in gradient descent?
It represents the direction of steepest ascent of a function at a point.
What does CNN stand for?
Convolutional neural network
Compute the angle between <1, -2> and <2, 4>.
arccos(-3/5)
What were our (Evan's and Albon's) Michigan Hackers projects?
Evan: reinforcement learning bot for BS
Albon: generating artwork from music
What kind of tokenization is used in GPT?
What does SVM stand for
Support-vector machine
What does the following series evaluate to?
sum_{i = 1}^infty 1/i^2
pi^2/6
When was Michigan Hackers founded?
2015
What do activation functions do? Why do we use them? Give two examples of activation functions.
Activation functions introduce nonlinearity to neural networks; we use them to capture complex, nonlinear relationships. Examples include ReLU, logistic, RBF, etc.
What does RBF stand for?
Radial basis function
What's Evan's favorite number?
856