l24_nr_ml

l24_nr_ml - Some Ideas From Machine Learning Slides adapted...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
i i Ni l / Session 24 Sli t & i ll Some Ideas From Mach ne Learn ng cho as Roy 16.410 13 des adap ed from: Andrew Moore Tom M tche , CMU Machine Learning Learning = improving with experience Improve over task T (e.g, Classification, control tasks) with respect to performance measure P (e.g., accuracy, speed, etc.) based on experience E (direct, indirect, teacher-provided, from exploration). What can we learn? Discrete classifications (a.k.a. discriminative classifiers) Probability of classifications (a.k.a. generative classifiers) Continuous Functions (a.k.a. regression) Control policies Model parameters You have already seen some of these ideas in Baum-Welch learning of Hidden Markov Models and Q-Learning in Reinforcement Learning 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Notation Instances: x 1 , x 2 , . .. Target concept C labels instance x with label c(x) Labelled data D is a set of pairs x, c(x) Hypothesis h is a possible target concept that also labels (correctly or incorrectly) each instance x with a label h(x) The inductive learning hypothesis Any hypothesis found to approximate the target function well over a sufficiently large set of training examples will also approximate the target function well over other unobserved examples. A hypothesis h is consistent with a set of training examples D of target concept c if and only if h(x) = c(x) for each training example x, c(x) in D . Consistent (h, D) ( ∀〈 x, c(x) 〉∈ D) h(x) = c(x) The version space , VS H,D , with respect to hypothesis space H and training examples D , is the subset of hypotheses from H consistent with all training examples in D . VS H,D {h H | Consistent(h, D)} 2
Background image of page 2
i i is is i i i l li li l ifi l l l + + + + + + + - - - - - - - - l + + + + + + + - - - - - - - - + i l l i l i Li i in li it’ ll ll ini l The Assumpt on of the Induct ve Hypothes We assume the target concept
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 9

l24_nr_ml - Some Ideas From Machine Learning Slides adapted...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online