2019 FEE 10 ASA / 2020 AEBE 4 BASE RRARE RECA EOI CLRE MARE AER GEFa#B) 2019 48 8 A 5 A 12:00~14:00 t 1, MRF Oded 13 Hs. 2. RRBRADAM ASE THER TILT. 3, BUERDROAU, HOU TRREL, ET RR MLANBIO RAE O DST UL ICPLIWS IE, 4, PREIS F526 BE CdD, HAGEL SERBOM AS CHRSH TW VS, LODHI 2 RRESRL, BPEL At ay, REAR, SO RBA S-2 RTE: 83 AIRE ER S-4 {HRs SS (25ue S-6 BABA, HEM MERE 5. ACHR D ARV RD, HAGEETSREE CHET OS, 6. APSARA CSE RII OM CLM ee The Japanese version of this document isthe prevailing and authoritative version; the English translation below is provided for reference only October 2019 Admissions / April 2020 Admissions Entrance Examination for Master’s Program Department of Intelligence Science and Technology Graduate School of Informatics, Kyoto University (Specialized Subjects) August 5, 2019 12:00 - 14:00 NOTES 1. This s the Question Booklet in 13 pages including this front cover Do not open the booklet until you are instructed to start ‘fier the examination has started, check the number of pages and notify proctors (professor) immediately if ‘you find missing pages or unclear printings. 4, There are 6 questions, waitten in Japanese and English. The questions are clasified as listed below. Choose and answer 2 questions. S-1 Cognitive Neuroscience, Cognitive and Perceptual Psychology “~~~ Pages 1 to2 S2 Statistics Pages 3 t0.4 S-3_ Pattern Recognition, Machine Leaming Pages 5to 6 S-4 Information Theory Pages 7t0 8 S-5_ Signal Processing: Pages 90 10 $-6 Formal Language, Theory of Computation, Discrete Mathematics + Pages 11 to 12 5. Write your answer in Japanese or English, unless otherwise specified. 6 — Read carefully the notes on the Answer Sheets as well18H SAHA (BARBS NH-BNDES) | MBBS si BAL FOR BICOMCMRCBREL, BEATE RMS (1) SS5RAERERVALK (transcranial magnetic stimulation: TMS) (2) ABELIAN (Petinotopy) Q) Pe—7— + 7x F— OPEB (Weber-Fechner law) (4) EASY 7 (visual masking) (5) 2 b— TAR (the Stroop effect) BH 2 WKA-ADEROMMEMASEOID 1 OL 5 2RME VERB EBL ke EAP) HH (Bla) 8 100ms RRENKOL, F—F y dish (Bib) ARS A D, RBSMH ORB, Mle OT DS T4 CRANE 4 SHOMBOV TAMIA S DEIAFY BX PODWTHP THEME, THENOF— 7 y bRFICHY ST SREAKEMP LE LOCMET SLE CHS, FOF y bARORRBA Se PLE COMOMME ROEM & LCRMETS, RRSMH, RTP, GHLOTF (+) SRMTS LI MRSS, () COMMS CHA ADERAHK ETS E. 4 BBO F—F y KT, 12, T3, TAO RISMMULL OL SICRSLPMSNSD, FRSNOBRETF7ERBEAT RMU L. HS, TIE TS ORO BIG LORE, TLE TA ORO BIH LOMAS LOD DET. Q) LROBRORME, HIS AORECBW THREL, (a) a) (b) B—Fobk BL. RATA SKF) Him @) - 2-H y bB Ob) OBIE, 2-H y MERLE (0Master's Specialized [Cognitive Neuroscience, Question Program Subjects Cognitive and Perceptual Psychology} |_ Number | 5? Question is translated in English in the section below; this translation is given for reference only. QI Give a brief explanation on each of the following items. Figures may be used. (1) Transcranial magnetic stimulation (TMS) (2) Retinotopy (3) Weber-Pechner law (4) Visual masking, (5) The Stroop effect Q2 For investigating the function of object-based attention, the following experiment was designed. As shown in Figure 1, a cue display (Figure 1a) is presented for 100 ms, followed by a target display (Figure 1b). A target (either X or O) is presented at one of the four locations (Tl, 12, 73, and T4 in Figure 1¢). The task of participants is to discriminate the target by pressing one of the two keys assigned to the target letters. The time from the onset of a target display to pressing, a key is measured as a reaction time. Participants are instructed to maintain fixation on the central cross on the displays throughout the trial (1) If object-based attention works in the task, what is your prediction about the reaction times for the four target locations (T1, T2, T3, and T4)? Explain your prediction in words and a graph. Note that the distance between T! and 73 is equivalent to that between TI and T4, (2) Discuss the reason of your predicted result above in terms of the function of object-based attention, (c) Target locations 14 Figure 1. Examples of a cue display (a) and a target display (b) used in the experiment. (c) Potential target locations.St SMAA (ete) | BS | s2 BL MABE X n= 4,p=0S OMMBAMICHEI LTS, COLE. Xa2E OMB P(X =2) OER, BM2 S524 ORL EMM SMB RHSPEIDERELLVY, CO2 Ave SHWE e GRATE, BEA BIE TS, BRITA 2S BWI LeRARRBE LELE, KOMMOPM (HORE) ERDK. BRAS ABOKM @ = 0.05 Cn MOWMARRERET SR, TTI CHR GRRE LICR DDDOT, DIRK Lb 1 DOMMRRERAM LT LED MBER ok BE 4 FY p Main, AHO 0? = 10 DEMDHP SIT AS ORREHH WILE ELA, WS 12 Choke. MMIERA AICHE 5 MBER Y [COWT, P(-1.96 SY $1.96) = 0.95 #RY MOL LEA LT, pO SWEBAMERD de BAS RDS RMLIERBE. HLA SRAM CEMA THODO EL 5 Cie “+ S¢FHSl4, HARKing (Hypothesizing After the Results are Known) WFIEIVS. BtEHRIE (CESK MRICBU YT MARKing MAM CH SMHS, HERE AV CRA deMaster's Program ‘Question is translated in English in the section below; this translation is given for reference only. Specialized Subjects [Statist Q1_ Consider a random variable X that follows the binomial distribution with m = 4 and p = 0.5. Compute the value of the probability P(X = 2), Q2 Consider testing whether there is a bias in the probabilities of the head and tail in a coin toss. Suppose that 5 independent trials of the coin toss resulted in 1 head and 4 tails. Compute the P-value of this observation (two-sided test), given the null hypothesis that there is no bias. Q3- Consider testing m independent hypotheses with the significance level @ = 0.05. Find the probability of rejecting at least one null hypothesis when all the null hypotheses are correct. QA Suppose that a sample of size 5 from the normal distribution with unknown mean 4. and known variance a? = 10 had a mean of 12. Estimate the 95% confidence interval of 4 using the fact that P(—1.96 < ¥ < 1.96) = 0.95 where ¥ follows the standard normal distribution. Q.5. The practice of presenting a hypothesis informed by results as if it were an a priori hypothesis is termed HARKing (Hypothesizing After the Results are Known). Explain, using statistical terms, why HARKing, inappropriate in studies based on statistical testing,S1BR SMHS (is-viememme mR) | Mees | so BEL DREN bY XEADEL, 27 7RACAMTOMBESAS, KTS HBA g@)ewiy CLE LEMME EGS) DEDE RZ IAL ECRWE RPIAVERBE SRM wW EEBES, ERIM y (0,1) ERE N OF BFW, Cm MIBERONTI Bo QOH L TLS MER (AF 77H) EMAL DORBM AT boy LOPES, NT hay PREC Sw ORRRERE, gO L CRY YE4 KBREMA LE BOMAUAT 4 y 7 REPENS, LOKERE, G) LRELYAF 4 77 AROMA AIL FA 1 OFAMBEBRES. ee BX, TIAL OWL 7 FA 0 OBER HOMRILMNS CLEAR. G) BYAT 4 oT BO ROMBAIL, —LDdylog f(x) + (1 — yp log — fe) CHENS, COS PIN LS w ORBRERE, EE (ON-ET h VeRORRAL OMREMAL, (6) SORT bE Yd SWEAT 1 7 RESET L, NEBR ROLL SBA-ET bAYRW UIT 4 —Faa—-F Ay bI—7(DNN) DH CE 5. LNEFET SLOORBURE (Sy 7 TONY av) EO PRBS K, Kit, LOPWRANAA CS SLOORMH, RULE VME MEAS EMCSRVBHER e(GIS]) EMEF viv) € SMRETSLE, fECORMMBE LED. CREAMBEEEORNO k eG ORMBAME RY, L(G) EBX. AED G1, G2, Gs, Cy BACHE NS IF 7 THY, Gy ¥ Gz ORAAOLSAAALERT, @-® bo Gy G Q Gs Gy () FF 7G BEUGDENENKWT, 4 PEMD HS PABAK, (2) FF 7 Gy DBE x(Gs) DMEBAL, HH, Gy cH &, BEI, MSABIC Cy EMBL, TOSM KY SAL, (v) DEBATES LET 8) FF 7 Gy OBAREML(Gs) DHEEAL, Eh. GATS L(G) MBSE 12, ANP (2) EAU BIRCHRA L, (4) XG) A LG) PROMO LS RAMBO ITE 1 TMREL, RK, THAT XG) # LG) SHR LTO SRAM, ABOU Cm 5 MARS CT Bu, (9) GB VEROIS 7 GILT X(G) < L(G) BMY WF. KIEL? SAL Kh EPELOPELS SUP EMRL, BOTT OMHEIRA L, ©) Ky EnWRREII7ETS, nS L(Kq) S In-1 CHES LEME, 11Master’s Specialized [Formal Language, Theory of Computation, { Question Program Subjects Discrete Mathematics} Number (Question is translated in English n the section below; this translation is given for reference only. s-6 Q. In this question, we consider only simple undirected graphs having at least one vertex. For a positive integer k and a graph @ = (V, B), let f be a mapping from the set of vertices V to (1,2...) If fu) # Flos) holds for any edge (v;,0;) € B, f is called a k-coloring of G ‘The smallest k such that G has 2 k-coloring is called the chromatic number of G and denoted by x(G). For $ C V, G[S} denotes the subgraph of G induced by S, and e(G[S}) denotes the number of edges in G[S]. If, for any $ C V (2 < [S| < 3), there exist u,v) € S such that [F(e:) ~ f(ej)| > e(G[S]), then f is called a local k-coloring of G. The smallest & such that has a local k-coloring is called the local chromatic number of G and denoted by L(G). The following Gy, Go, Gs, and Gs are graphs used in the questions, where the labels of vertices in Gx and Go represent their names, Gy Gs Gs (1) For each of Gi and Gp, give the number of 4-colorings. (2) Give the chromatic number x(@3) of Gs. Also, give a x(Gs)-coloring of G3. For giving a coloring, draw G5 on your answer sheet, and write the value of f(v) inside each vertex v. 3) Give the local chromatic number L(G) of Ga. Also, give a local L(G«)-coloring of Gs For giving a local coloring, do the same as in question (2), (4) Draw a graph G with the minimum number of vertices that satisfies x(G) # L(G). Also, explain why your answer satisfies x(G) # L(G). You do not need to give a reason for the ‘number of vertices being minimum, (5) Isthe proposition “For any graph G, x(@) < L(G) holds.” correct? To answer this question, first mention whether it is correct or not, and subsequently give @ reason. (6) Let Ky be the complete graph with n vertices. Show that n < L(K,) < 2n— 1 12