Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
CSE 531: Performance Analysis of Systems Lecture 2: Probs & Stats review Anshul Gandhi 1307, CS building [email protected] [email protected] 1 Outline 1. Announcements 2. Probability basics Experiments, events, helpful relations 3. Random variables Discrete Bernoulli, Binomial, Geometric Continuous Uniform, Exponential 2 Announcements • Collaborating on assignments • Assignment 1 (next week) 3 Basics • • • Probability is defined in terms of some experiment. The set of all outcomes of an experiment is its sample space. A subset of the sample space is called an event. Mutually exclusive Partition Independent • A function defined on the outcomes is a random variable. • • • Law of total probability Conditional probability Bayes’ theorem 4 Random variables • Discrete and Continuous • Discrete Countable possibilities pmf 5 Discrete RVs • PMF for sample space S Pr[X = s] = pX(s) = p(s) p( s) 1 sS CDF: FX(a) = Pr[X ≤ a] = Pr[ X a] xa Inverse CDF: F̅X(a) = Pr[X > a] = 1 - FX(a) = Pr[ X a] xa Mean E[X] = s. p ( s ) sS E[X2] 2 s = . p( s) sS Var[X] = E[X2] – (E[X])2 6 Bernoulli(p) • Outcome of a coin toss • p(1) = p • p(0) = 1-p p( s) 1 sS (find limits of s) Mean E[X] E[X2] Var[X] 7 Binomial(n, p) • Number of 1’s when flipping a Bernoulli coin n times • p(i) = nCi pi (1-p)(n-i) p( s) 1 sS Mean E[X] E[X2] Var[X] 8 Geometric(p) • Number of flips till we get a 1 • p(i) = (1-p)(i-1) . p p( s) 1 sS Mean E[X] E[X2] Var[X] 9 Continuous RVs • PDF for sample space S b b a a Pr[a ≤ X ≤ b] = f X ( x)dx f ( x)dx f ( x) dx 1, f ( x) dx Pr[ x X x dx] a CDF: FX(a) = Pr[X ≤ a] = f ( x ) dx f ( x) d dx F ( x) a E[Xi] i x = f ( x ) dx Var[X] = E[X2] – (E[X])2 10 Uniform(a, b) • f(x) = 1/(b-a) for a < x < b f ( x)dx 1 E[X] E[X2] Var[X] 11 Exponential(λ) • f(x) = λ e - λ x, x ≥ 0 f ( x)dx 1 E[X] E[X2] Var[X] 12