1 / 91

Pattern Recognition and Machine Learning

Pattern Recognition and Machine Learning. Chapter 2: Probability distributions. Two approached to density estimation. Parametric Non Parametric. 1. Parametric Estimation of Density. Basic building blocks: Need to determine given Representation: or ?

ilyssa
Download Presentation

Pattern Recognition and Machine Learning

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Pattern Recognition and Machine Learning Chapter 2: Probability distributions

  2. Twoapproachedtodensityestimation Parametric NonParametric

  3. 1. Parametric Estimation of Density Basic building blocks: Need to determine given Representation: or ? Recall Curve Fitting

  4. Assumptions 1. There is a known parametric form 2. Data is i.i.d. Note: given data you can fit infinitely many p(x)

  5. Binary Variables (1) Exp 1: Coin flipping: heads=1, tails=0 Bernoulli Distribution

  6. Binary Variables (2) Exp 2:N coin flips: Binomial Distribution

  7. Binomial Distribution

  8. Parameter Estimation (1) MLE for Bernoulli Given:

  9. Parameter Estimation (2) Example: Prediction: all future tosses will land heads up Overfitting to D

  10. CongugatePrior RecallBayesformula: p(w|x, t) ∝ p(t|x, w)p(w). Assume, a functional form of the prior is the same as the functional form of posterior. Need to find cojugate prior of the likelihood

  11. Beta Distribution Distribution over .

  12. Gamma Function: Extension of FactorialFunction

  13. Beta Distribution

  14. Bayesian Bernoulli The Beta distribution provides the conjugate prior for the Bernoulli distribution.

  15. Prior ∙ Likelihood = Posterior

  16. Properties of the Posterior As the size of the data set, N , increase

  17. Sequential Approach for Density Estimation Suppose we receive observation one at a time. What is the probability that the next coin toss will land heads up?

  18. Multinomial Variables 1-of-K coding scheme:

  19. LagrangeMultipliers maximize f(x,y) subject to g(x,y)=c Define andmaximize

  20. ML Parameter estimation Given: Ensure , use a Lagrange multiplier, ¸.

  21. The Multinomial Distribution

  22. The Dirichlet Distribution Simplex in K dimension: Conjugate prior for the multinomial distribution.

  23. Bayesian Multinomial (1) Conjugate prior of Multinomial is Dirichlet distribution

  24. Bayesian Multinomial (2)

  25. The King of the Distributions: Gaussian

  26. Central Limit Theorem The distribution of the sum of Ni.i.d. random variables becomes increasingly Gaussian as N grows. Example: N uniform [0,1] random variables.

  27. Geometry of the Multivariate Gaussian

  28. Moments of the Multivariate Gaussian (1) thanks to anti-symmetry of z

  29. Moments of the Multivariate Gaussian (2)

  30. Partitioned Gaussian Distributions PartitiontheGaussian data X intotwoparts: What is thedistribution of eachpart? P(xa) What is theconditionaldistributions? P(xa/xb)

  31. Partitioned Conditionals and Marginals:TheyareallGaussian…

  32. Partitioned Conditionals and Marginals

  33. Bayes’ Theorem for Gaussian Variables Given we have where

  34. Maximum Likelihood for the Gaussian (1) Given i.i.d. data , the log likeli-hood function is given by Sufficient statistics

  35. Maximum Likelihood for the Gaussian (2) Set the derivative of the log likelihood function to zero, and solve to obtain Similarly

  36. Maximum Likelihood for the Gaussian (3) Under the true distribution Hence define

  37. Sequential Estimation Contribution of the Nth data point, xN correction given xN correction weight old estimate

  38. The Robbins-Monro Algorithm (1) Consider µ and z governed by p(z,µ) and define the regression function • Seek µ? such that f(µ?) = 0.

  39. The Robbins-Monro Algorithm (2) Assume we are given samples from p(z,µ), one at the time.

  40. The Robbins-Monro Algorithm (3) • Successive estimates of µ? are then given by • Conditions on aN for convergence :

  41. Robbins-Monro for Maximum Likelihood (1) Regarding as a regression function, finding its root is equivalent to finding the maximum likelihood solution µML. Thus

  42. Robbins-Monro for Maximum Likelihood (2) Example: estimate the mean of a Gaussian. The distribution of z is Gaussian with mean ¹ { ¹ML. For the Robbins-Monro update equation, aN = ¾2=N.

  43. Bayesian Inference for the Gaussian (1) Assume ¾2 is known. Given i.i.d. data , the likelihood function for¹ is given by This has a Gaussian shape as a function of ¹ (but it is not a distribution over ¹).

  44. Bayesian Inference for the Gaussian (2) Combined with a Gaussian prior over ¹, this gives the posterior Completing the square over ¹, we see that

  45. Bayesian Inference for the Gaussian (3) … where Note:

  46. Bayesian Inference for the Gaussian (4) Example: for N = 0, 1, 2 and 10.

  47. Bayesian Inference for the Gaussian (5) Sequential Estimation The posterior obtained after observing N { 1 data points becomes the prior when we observe the Nth data point.

  48. Bayesian Inference for the Gaussian (6) Now assume ¹ is known. The likelihood function for ¸ = 1/¾2 is given by This has a Gamma shape as a function of ¸.

  49. Bayesian Inference for the Gaussian (7) The Gamma distribution

  50. Bayesian Inference for the Gaussian (8) Now we combine a Gamma prior, ,with the likelihood function for ¸ to obtain which we recognize as with

More Related