filmov
tv
04: Real Learning Is Feasible (100min)
Показать описание
Machine Learning From Data, Rensselaer Fall 2020.
Professor Malik Magdon-Ismail talks about real learning, not verification. We discuss how choosing a hypothesis from a hypothesis set that is fixed before you see the data poses problems for the Hoeffding bound. We can modify the Hoeffding bound to account for the ability to select a hypothesis from H, and the price that must be paid is exactly the size of the hypothesis set. This brings us to the two step process for real learninig: (1) Get Eout=Ein and (2) Get Ein=0. The theory holds for any target function but we can now argue that more complex target functions are harder to learn because they need more data. Lastly we bring our learning setup closer to practical settings by discussing noisy target functions and choice of error/risk measures.
This is the fourth lecture in a "theory" course focusing on the foundations of learning, as well as some of the more advanced techniques like support vector machines and neural (deep) networks that are used in practice.
Level of the course: Advanced undergraduate, beginning graduate. Knowledge of probability, linear algebra, and calculus is helpful.
Professor Malik Magdon-Ismail talks about real learning, not verification. We discuss how choosing a hypothesis from a hypothesis set that is fixed before you see the data poses problems for the Hoeffding bound. We can modify the Hoeffding bound to account for the ability to select a hypothesis from H, and the price that must be paid is exactly the size of the hypothesis set. This brings us to the two step process for real learninig: (1) Get Eout=Ein and (2) Get Ein=0. The theory holds for any target function but we can now argue that more complex target functions are harder to learn because they need more data. Lastly we bring our learning setup closer to practical settings by discussing noisy target functions and choice of error/risk measures.
This is the fourth lecture in a "theory" course focusing on the foundations of learning, as well as some of the more advanced techniques like support vector machines and neural (deep) networks that are used in practice.
Level of the course: Advanced undergraduate, beginning graduate. Knowledge of probability, linear algebra, and calculus is helpful.
04: Real Learning Is Feasible (100min)
03: Is Learning Feasible? (102min)
02.pre.04 « Feasibility of Learning « Machine Learning « NUS School of Computing
04: Five Questions on Feasibility of Learning, Supplement (19min)
#machinelearning Is Learning Feasible? Hoeffding's inequality || ML- lec-05 || Learn With Debj...
04-a LFD: Hoeffding bound for real learning.
Is It Really Possible to Learn to Speed Read?
becoming smart is easy, actually
03-b LFD: Learning is feasible in the bin model, but not for sure. The Hoeffding Bound.
04-f LFD: Full learning setup - it must be possible to fail.
04-c LFD: Fit and Predict. The 2-step approach to learning.
How to learn code as fast as humanly possible
Learn these as quickly possible 🫶
Lecture 4: The Feasibility of (Machine) Learning
Learn How To Play Guitar Fast: Beginner Tips, Easy Riffs and First Chords
Effective study techniques : How Learning Happens - Session 04
How I'd Learn AI in 2024 (if I could start over)
Learn THIS As Quickly As Possible
Learn Python with ChatGPT: is this possible?
04-d LFD: Complex and noisy target functions.
Learning Together PSA Kim Possible
Is it possible to Learn an Architecture Program by Yourself?
#SHAPELive | How can we create real learning with technology?
RSS 2021, Spotlight Talk 14: Active Learning of Abstract Plan Feasibility
Комментарии