Lecture 18 – April 11 , 2016
نویسندگان
چکیده
An easy example of a concept class is the set of emails and functions mapping emails to spam or not spam. Another example is the set of points and a line `, and a single function determining which side of the line ` the points are on. PAC learning is the following problem. We are given a concept class H, a function f ∈ H, and a hidden distribution D on X. The algorithm is allowed to get m labeled examples (xi, f(xi)) for 1 ≤ i ≤ m, by drawing each xi ∈ X according to the distribution D. We want that for any constants , δ, after getting these m examples, with probability 1 − δ, the algorithm should ensure that the error on future examples drawn from the distribution D is ≤ . The δ denotes the probably, and the denotes the approximately in the name PAC learning. Our final definition here will be a weak learner. Definition 2. A weak learner is one that has error at most 1 2 − η for some η > 0.
منابع مشابه
Math 140a: Foundations of Real Analysis I
1. Ordered Sets, Ordered Fields, and Completeness 1 1.1. Lecture 1: January 5, 2016 1 1.2. Lecture 2: January 7, 2016 4 1.3. Lecture 3: January 11, 2016 7 1.4. Lecture 4: January 14, 2014 9 2. Sequences and Limits 13 2.1. Lecture 5: January 19, 2016 13 2.2. Lecture 6: January 21, 2016 15 2.3. Lecture 7: January 26, 2016 18 2.4. Lecture 8: January 28, 2016 21 3. Extensions of R: the Extended Rea...
متن کاملSecurity Protocols XVI - 16th International Workshop, Cambridge, UK, April 16-18, 2008. Revised Selected Papers
Why should wait for some days to get or receive the security protocols xvi 16th international workshop cambridge uk april 16 18 2008 revised selected papers lecture notes in computer science book that you order? Why should you take it if you can get the faster one? You can find the same book that you order right here. This is it the book that you can receive directly after purchasing. This secu...
متن کاملThe Midwife in India *Being a lecture delivered on 27th April 1925 during "Baby Week" at Belgaum.
متن کامل
Lecture 21 – April 25 , 2016
In this lecture we’ll explore Random Matrix Theory by exploring an application in network analysis. In particular, we’ll examine how Random Matrix Theory can be employed to develop an efficient randomized algorithm for finding planted cliques. In doing so, we will explore a general method for analyzing random graph problems with “planted” structure by treating them as a signal reconstruction pr...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2016