Learning From Data
Yaser Abu-Mostafa's famed Caltech course on statistical learning theory: Hoeffding's inequality, VC dimension, bias-variance tradeoff, regularization, support vector machines, and neural nets.
Complete university courses recorded at Caltech.
Yaser Abu-Mostafa's famed Caltech course on statistical learning theory: Hoeffding's inequality, VC dimension, bias-variance tradeoff, regularization, support vector machines, and neural nets.
Self-contained seminars, public lectures, and invited talks.
Connects Boltzmann's statistical definition of entropy to cosmology's deepest puzzle: why the early universe began in an extraordinarily low-entropy gravitational state.
Recounts half a century of geometrodynamics, numerical relativity, and laser interferometry culminating in LIGO's direct observation of colliding binary black holes.
VLSI and neuromorphic engineering pioneer Carver Mead reflects on device scaling limits, electron tunneling, Moore's Law, and physical laws governing computation.
Maps the physical boundary between noisy intermediate-scale quantum devices and fault-tolerant topological error correction.
An in-depth analysis of the feasibility of worst-case membership inference in large language models, challenging prevailing assumptions about data privacy.
An examination of privacy implications in Large Language Models, extending beyond memorization to reconstructive adversarial techniques and data poisoning vulnerabilities.



Physical law, LIGO gravitational waves, singularity topology, and AdS/CFT entanglement
Cornell University · Caltech · University of Oxford +2



Linear geometry, convex duality, statistical learning, and deep optimization
MIT · Stanford University · Caltech +1
Browse every recorded lecture from Caltech.