The ability of recently engineered machine learning algorithms and natural systems to perform inference and generalize well from limited finite data observations poses interesting challenges and open problems. This seminar will discuss both practical algorithms for applications and related rigorous mathematical theory. Examples include the approximation and generative abilities of deep learning with recent types of neural networks, formulations and training of unsupervised methods such as transformers, diffusion-models, autoencoders, and non-neural network approaches such as support vector machines, kernel methods, and probabilistic methods.
1
UnitsPass no pass
Grading1, 2, 3
PasstimeInter collegiate athletes only
Level LimitLetters and science
CollegeVery good professor. The lecture is little bit boring but he posted very informative notes on his website. The practice exam was similar to the real exams. Even though you get the wrong answer on the exam , he gave 90% credits for the steps.
Horrible class and big lack of support in all aspects.
He has a tough vocabulary in his lectures that makes you not be able to understand what's happening (he says "intuitively" 2932741 times), especially when class is early in the morning its hard to want to go to class. Only reason I have an A is because I crammed for the midterm and learned everything reading the textbook chapters.
Such a genuine human and a great lecturer was very easy too
In my opinion, professor Atzberger is not a bad lecturer. He is definitely passionate and knowledgeable about what he is teaching, but he forgot that we are all just undergraduate students trying to pass, not PhD students doing research with him. The exams are super easy though, the first time I got 100/100 in a college math course.
Godly Professor, cares about students wellbeing and adjusts his class based on students needs.