Introduction to Machine Learning Week 1 - Live Session Raj V Jain (PhD Student in IISc) PMRF TA Today's Agenda: Some Probability and Linear Algebra Problems.Administrative Things 1. Whatever is written will be made available in PDF format & session is recorded - so focus on solving rather than note taking 2. We are 4 PMRF TAs - My session will be on Saturdays from 4-5 PM 3. Welll usually do some problem solving, programming & doubt clearing session 4. Unable to see chat - feel free to interrupt me and ask doubt 5. In case you or | get dropped because of exceeding time, just join again3 ale [Link] ote a Jal contains k — J blue balls and n ~ K red balls. You pick a bin at random and remove two balls at random without ¥ Find the probabilty that: ‘+ tne second ball is red; + the second ball is red, given that the frst is red. There are n bins o O1r, 28 hich the k th contains k-1 blue balls and n-k red 1,13 ind remove two balls at random without O1n,28 lacement. Find the probability that @mnaD the second ball is red ie second ball is red, given that the first is red —e—e—e—ee—e—e—eeeeeeese__r—ese \ Opce a DLL. tS ible Pm [' Keo red G2 ral O Whe & \e-| blues Le uu nol ned Wk ged Loedbs ARPoreTotad Prtoodatiy oe - Cc, Cy Pw=5 "ey plo= plplaP te) & Uy v2) + PLDIC2) Kee) QA =a UVGQssYa) = S Ve\Ee) Peg) = i 5 We ye\ NW wu n-| a ol 2 o-) nory) pe me ma: 225 oth CalfPlato Piney | PasBrctrt\v dy ot re P (pe. =QRr | Rermrct teh i aed Lh ques Be Kar Jato Pink na 28 a Jah! PLU) Z > PCR) By) PCy) potato i> wo) P(e) Do mle Nees | = 3 woh ne , 1D 2 > Wel) \r\ v2) POY)= Ss (mem et > — wlrn-))\n-2) ye\ 2 LS GPE nE en ene oe] n(n-VXn-2) 2 os \ 3B pe nolenm) 6 b>) G V4 2A 2 \l + z 1 z ee7. Let A™*” be a matrix of real numbers. The matri 7 has an eigenvector x with eigenvalue b, Then the ei is f ATA which has eigenvalue b is equal to _b, Then the cigenvee ony w as eigenvalue b is equi a Bycang Ap! ae bx ‘a 4 x + 7 Cannot be described in terms of x a Aleta] Hi 3 ‘t (0) (4) S9. Let_u bean xtwector, such that uu =1. Let I be the nx n identity matrix. The n xn mairix A is given by (I — ku”), where is a real constant. w itself is an eigenvector of A, with eigenvalue -1. What is the value of k? @) 2 A- T-kuul (b) -1 oe ew Sep { Buck (a) 0 su - (Deeuui}w = Uke asuc-lku =k kusrou6) Let X and ¥ be a uniformly distributed random variable over the interval [0, 4] and (0, 6] respectively. i X 1 point and ¥ are independent events, then compute the probability, P(max(X, Y) > 3) 1 OO ie foe eye ¢ Oo oO roe. a PL wex Lay) 23) 5 = P( x>3, yer)+ PL <3, ¥>3) Et 4 Px >3, NEx®) 2 2 OCx>d) P¥E3) 4 Foxe PU) g SEVLK > BPLY DR) (PL ¥) = POD PLY) ae aur > +3yteapt = FScrv ( Flay= )— 062% rene 2 chet =p(x x) abe 0 Bem OXLO ° geo1. Given a bag containing 6 red balls, 4 blue balls and 7 green balls, what is the probability that in 5 trials, at least d balls are drawn from the bag? = Ly Replacement / 2. In the experiment from the previous question, what is the probability of y picking a red ball for the first time on the fourth attempt? oe & ( aed R= prcwex nef = 17 See Va wok prce-g vet nee Ax 22) vO Ww E Bo wk PUKaKd ="C) p Qed PUK23)= 3 axe a Ry 2sLayer dw prele = “s Creome “ hye Ahace Ak R =k Larne Seer Sek W2)= Wey (2) = WY 7 2(UI E> 9? (>>(Lvvrow h Q Cyeonreme, ak Qeasie Sie AS Rav s sed ) 2p eee Zeer roead Ler t ae Pe =Brainstorming session Ors DUAN Y Let: hi t dict t rs ther. e ets say you have to predict jomorrow’s weal ler. arid How would you go about it? Sum clory 0 hyd ox I, weatteY Z@) Yoav oyeor yectan Dury ey tinne | tes @ Wr dtp & dul _— t Olesug —] OSE}, Dwser > Coren ritySfooge UR © xo \ “4 7 oN) — eK | ton aS a 7 9 wn ® pee 7 WAY = = Qe ire Fabs ENN | ey HMM SeeAny Doubts?