Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of GraphSearch.
This lecture by the instructor delves into understanding machine learning through exactly solvable models, exploring topics such as sample complexity, models in data science and physics, neural network learning, and the Bayes-optimal generalization. The lecture also covers the teacher-student perceptron, the closed-form solutions, and the computational gaps in learning. The discussion extends to the committee machine, the Langevin state evolution, and the hidden manifold model, providing insights into the landscape analysis and the transition recipes in deep learning theory.