Slide 1

Slide 1 text

%FFQ/FVSBM/FUXPSLͷڞಉֶश ౻٢߂࿱ʢத෦େֶɾػց஌֮ϩϘςΟΫεάϧʔϓʣ IUUQNQSHKQ

Slide 2

Slide 2 text

೥݄ɿۚग़෢༤ઌੜ͔Β௖͍͓ͨݴ༿ 2 த෦େֶϩΰ த෦େֶϩΰ

Slide 3

Slide 3 text

w ʮൃ૝͸୯७ɼૉ௚ɼࣗ༝ɼ؆୯ で ͳ͚Ε ば ͳΒͳ͍ɽ͔͠͠ɼൃ૝Λ࣮ߦʹҠ͢ʹ͸஌ࣝ が ͍Δɼख़࿅͞Εٕͨ が ͍Δʯʢۚग़෢༤ʣ ʮૉਓൃ૝ݰਓ࣮ߦʯͱ͸ʁ 3 த෦େֶϩΰ த෦େֶϩΰ IUUQTXXXBNB[PODPKQEQ#11428/ IUUQNQSHKQ5,CPPL

Slide 4

Slide 4 text

w ʮݴ͏͸қ͘ɼߦ͏͸೉͠ʯͷయܕ w ೥͔Β໿೥Λܦͯʮૉਓൃ૝ݰਓ࣮ߦʯ͚ۙͮͨݚڀ ඇ౳ํੑ-P(ϑΟϧλʹΑΔෳ਺ͷΞϑΟϯྖҬͷਪఆ<)BTFHBXB *$$7> ஌ࣝసҠ グ ϥϑʹΑΔڞಉֶश<.JOBNJ "$$7> <0LBNPUP &$$7> ݚڀϞοτʔʮૉਓൃ૝ݰਓ࣮ߦʯ 4 த෦େֶϩΰ த෦େֶϩΰ ࿦จ" ࿦จ# ࿦จ$ ࿦จ% ࿦จ" ࿦จ# ࿦จ$ ࿦จ% ࿦จ& ଟ͘ͷ࿦จΛਂ͘ಡΜͰ͍͘ͱ ຊ࣭తͳݚڀ ຊ࣭Ͱͳ͘খ͞ͳ͜ͱ ʹண໨ͨ͠ݚڀ ݚڀۭؒ

Slide 5

Slide 5 text

ਂ૚ֶशͷωοτϫʔΫߏ଄ த෦େֶϩΰ த෦େֶϩΰ w *-473$ʹ͓͚ΔωοτϫʔΫߏ଄ͷมભ 2012 SuperVision GoogLeNet Konvolüzasyon Pooling Softmax Diğer [Krizhevsky NIPS 2012] [Szegedy arxiv 2014]-22 [Sim "MFY/FU MBZFST *-473$ 2014 GoogLeNet Konvolüzasyon Pooling Softmax Diğer VGG MSRA [Szegedy arxiv 2014]-22 [Simonyan arxiv 2014] -19 [He arxiv 2014] n 2014 GoogLeNet Konvolüzasyon Pooling Softmax Diğer VGG MSRA 35/36 t derin öğ kullanm 20/36 t açık-kay Caffe uygula kullanm 012] [Szegedy arxiv 2014]-22 [Simonyan arxiv 2014] -19 [He arxiv 2014] 7(( MBZFST *-473$ (PPHMF/FU MBZFST *-473$ 14 3FT/FU MBZFST *-473$ ˠ໿ ສύϥϝʔλ ˠ໿ԯύϥϝʔλ ˠ໿ ສύϥϝʔλ

Slide 6

Slide 6 text

,OPXMFEHF%JTUJMMBUJPO ,% <()JOUPO > த෦େֶϩΰ த෦େֶϩΰ w ஌ࣝৠཹ ֶशࡁΈ-BSHFωοτϫʔΫ͔Β4NBMMωοτϫʔΫʹ஌ࣝసҠ ੑೳΛอͪͭͭɺύϥϝʔλ਺ͱܭࢉίετΛ࡟ݮՄೳ 5FBDIFS /FUXPSL 4UVEFOU /FUXPSL -BSHF QSFUSBJOFE 4NBMM ,OPXMFEHFUSBOTGFS ஌ࣝৠཹ ஌ࣝ

Slide 7

Slide 7 text

,OPXMFEHF%JTUJMMBUJPO ,% <()JOUPO > த෦େֶϩΰ த෦େֶϩΰ w ஌ࣝৠཹɿ5FBDIFSˠ4UVEFOU ֶशࡁΈ-BSHFωοτϫʔΫ͔Β4NBMMωοτϫʔΫʹ஌ࣝసҠ ੑೳΛอͪͭͭɺύϥϝʔλ਺ͱܭࢉίετΛ࡟ݮՄೳ ֶशํ๏ɿ)BSEUBSHFUͱ4PGUUBSHFUͰ4UVEFOUωοτϫʔΫΛֶश %BSL,OPXMFEHFʢӅΕͨ஌ࣝʣ 5FBDIFS 4UVEFOU $SPTT&OUSPQZ $SPTT&OUSPQZ MBCFM QSFUSBJOFE 4PGUUBSHFU )BSEUBSHFU ʢਖ਼ղϥϕϧʣ p1 p2 ʢ֬཰෼෍ʣ #BDLQSPQ

Slide 8

Slide 8 text

%FFQ.VUVBM-FBSOJOH %.- <:;IBOH > த෦େֶϩΰ த෦େֶϩΰ w 4UVEFOUωοτϫʔΫͷ૬ޓֶशɿ4UVEFOU⁶4UVEFOU ,VMMCBDL-FJCMFS ,- %JWFSHFODFͰωοτϫʔΫͷग़ྗ ͱ Λ͚ۙͮ߹͏ p1 p2 MBCFM 4UVEFOU 4UVEFOU p1 p2 KL(p1 ||p2 ) KL(p2 ||p1 ) ˠྠߨͷΑ͏ͳֶशܗଶͰ͋Γɺ஌ࣝৠཹΑΓਫ਼౓͕޲্ )BSE5BSHFU $SPTT&OUSPQZ 4PGU5BSHFU ,-EJWFSHFODF

Slide 9

Slide 9 text

%FFQ.VUVBM-FBSOJOH %.- <:;IBOH > த෦େֶϩΰ த෦େֶϩΰ w Կނ%.-͸͏·͍͘͘ͷ͔ʁ *OEFQFOEFOU Y Zd Y Zd %.- 7*46"-*;*/(5)&-044-"/%4$"1&0'/&63"-/&54 ˠ%.-͸޿͍୩ʹམ͍ͪͯΔ

Slide 10

Slide 10 text

%FFQ.VUVBM-FBSOJOH %.- <:;IBOH > த෦େֶϩΰ த෦େֶϩΰ w ޿͍୩ʹམ͍ͪͯΔͱԿނྑ͍ͷ͔ʁ ύϥϝʔλ ʢ΋͘͠͸ೖྗ ʣ͕มಈͯ͠΋ଛࣦ͸มԽ͠ͳ͍ˠ൚Խೳྗ͕ߴ͍ *OEFQFOEFOUɿଛࣦ͕খ͘͞ͳΔ୩ʹऩଋʢ޿͍୩ͱ͸ݶΒͳ͍ʣ %.-ɿೋͭͷωοτϫʔΫͷग़ྗ͕ࣅΔΑ͏ʹֶशˠ޿͍୩Λ୳͢ ˠϩεͷϥϯυεέʔϓ͕ޯ഑ͷ؇΍͔ͳತؔ਺ʹ͖ۙͮਫ਼౓΋޲্ w x -PTT y = f(x) y = wTx w ࣸ૾ؔ਺ɿχϡʔϥϧωοτϫʔΫ " # %.- %.- *OEFQFOEFOU

Slide 11

Slide 11 text

ωοτϫʔΫؒͷ஌ࣝసҠ த෦େֶϩΰ த෦େֶϩΰ w ޿͍୩ʹམ͍ͪͯΔͱԿނྑ͍ͷ͔ʁ஌ࣝΛ఻͑Δ͜ͱͰೝࣝੑೳ͕޲্ ஌ࣝৠཹɿ,OPXMFEHF%JTUJMMBUJPO<()JOUPO > ૬ޓֶशɿ%FFQ.VUVBM-FBSOJOH<:;IBOH > 5FBDIFS 4UVEFOU ,OPXMFEHF%JTUJMMBUJPO ,% 4UVEFOU 4UVEFOU %FFQ.VUVBM-FBSOJOH %.- p1 p2 p1 p2

Slide 12

Slide 12 text

஌ࣝৠཹɾ૬ޓֶशͷ೿ੜख๏ த෦େֶϩΰ த෦େֶϩΰ 4UVEFOU 4UVEFOU 4UVEFOU 4UVEFOU 4UVEFOU 4UVEFOU 4UVEFOU 5FBDIFS 4UVEFOU 5FBDIFS 4UVEFOU 5FBDIFS 5" 4UVEFOU ,OPXMFEHF%JTUJMMBUJPO %FFQ.VUVBM-FBSOJOH ,OPXMFEHF%JTUJMMBUJPO <()JOUPO > #PSO"HBJO <'VSMBOFMMP > 5FBDIFS"TTJTUBOU <.JS[BEFI > %FFQ.VUVBM-FBSOJOH<:;IBOH > -BSHFϞσϧ4NBMMϞσϧ ಉҰαΠζ தؒϞσϧ͋Γ -BSHFϞσϧ4NBMMϞσϧ ಉҰαΠζ Ϟσϧ਺ p1 p2 p1 p2 p1 p2 p1 p2 p1 p2 p3 p1 p2 p3

Slide 13

Slide 13 text

஌ࣝৠཹɾ૬ޓֶशͷ೿ੜख๏ த෦େֶϩΰ த෦େֶϩΰ ,OPXMFEHF%JTUJMMBUJPO <()JOUPO > #PSO"HBJO <'VSMBOFMMP > 5FBDIFS"TTJTUBOU <.JS[BEFI > %FFQ.VUVBM-FBSOJOH<:;IBOH > ,OPXMFEHF%JTUJMMBUJPO %FFQ.VUVBM-FBSOJOH -BSHFϞσϧ4NBMMϞσϧ ಉҰαΠζ தؒϞσϧ͋Γ -BSHFϞσϧ4NBMMϞσϧ ಉҰαΠζ Ϟσϧ਺ ਓ͕ઃܭͨ͠ݶఆతͳֶशํ๏

Slide 14

Slide 14 text

ຊݚڀͷ໨ඪ த෦େֶϩΰ த෦େֶϩΰ w ڞಉֶशΛΫϥεϧʔϜεέʔϧ΁֦ு ଟ༷ੑͷߴ͍ڞಉֶशΛ࣮ݱ ʮૉਓൃ૝ʯ ˠڭࣨͰͷֶशͷΑ͏ʹઌੜ͔ΒͰͳ͘ଟ͘ͷੜె͕ෳࡶʹڭ͑͋͏ֶश

Slide 15

Slide 15 text

w ڞಉֶशΛΫϥεϧʔϜεέʔϧ΁֦ு ଟ༷ੑͷߴ͍ڞಉֶशΛ࣮ݱ w ஌ࣝసҠάϥϑͷఏҊ<.JOBNJ "$$7> ैདྷख๏ ,%ͱ%.- Λ಺แͭͭ͠ɺ৽ֶ͍͠शํ๏ΛؚΉදݱํ๏ ̐छྨͷ(BUFؔ਺ʹΑΓ஌ࣝసҠΛ੍ޚ͢Δ͜ͱͰଟ༷ͳڞಉֶश ,% %.- ຊݚڀͷ໨ඪ த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 ஌ࣝసҠάϥϑ

Slide 16

Slide 16 text

w άϥϑΛ༻͍ͯ,%ͱ%.-Λදݱ ϊʔυɿਂ૚ֶशϞσϧ Τοδɿ஌ࣝৠཹͷଛࣦ άϥϑදݱ΁ͷม׵ த෦େֶϩΰ த෦େֶϩΰ ,OPXMFEHF%JTUJMMBUJPO ,% 5FBDIFS 4UVEFOU 4UVEFOU 4UVEFOU %FFQ.VUVBM-FBSOJOH %.- p1 p2 p1 p2 -BSHF 4NBMM m1 m2 ํ޲ͷΤοδ -BSHF 4NBMM m1 m2 ૒ํ޲ͷΤοδ άϥϑදݱ

Slide 17

Slide 17 text

w ิॿϊʔυ͕ධՁର৅ϊʔυͷֶशΛαϙʔτ͢Δ ϊʔυɿਂ૚ֶशϞσϧ Τοδɿ஌ࣝৠཹͷଛࣦ ஌ࣝసҠάϥϑ ϊʔυ਺͕ͷ৔߹ த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 ਖ਼ղϥϕϧ ධՁର৅ϊʔυ 3FT/FU ิॿϊʔυ 3FT/FU 8JEF3FT/FU %FOTF/FU …

Slide 18

Slide 18 text

w ֤ΤοδʹҟͳΔଛࣦؔ਺Λఆٛ ଛࣦؔ਺ͷ૊Έ߹ΘͤΛ୳ࡧ͢Δ͜ͱͰ৽ͨͳֶशํ๏Λ࣮ݱ ஌ࣝసҠάϥϑ ϊʔυ਺͕ͷ৔߹ த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 ଛࣦؔ਺ 𝐿 = 𝐻 ( 𝑝 ^ 𝑦 , 𝑝 𝑛 ) 𝐿 = 𝐾 𝐿 ( 𝑝 𝑛 || 𝑝 𝑚 ) 𝐿 = 0 … ˠଟ༷ͳ஌ࣝసҠΛදݱ͢ΔϑϨʔϜϫʔΫΛઃܭ ʮݰਓ࣮ߦᶃʯ

Slide 19

Slide 19 text

w ϊʔυ̎ TPVSDF ͔Βϊʔυ̍ EFTUJOBUJPO ΁ͷ஌ࣝసҠ Τοδͷ஌ࣝసҠͷଛࣦܭࢉ த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝑚 1 𝑚 2 𝐿 2,1 -PTT GVOD 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 p2 (c|x) p1 (c|x) L2,1 (p2 , p1 ) 'PSXBSE

Slide 20

Slide 20 text

w ϊʔυ̎ TPVSDF ͔Βϊʔυ̍ EFTUJOBUJPO ΁ͷ஌ࣝసҠ Τοδͷ஌ࣝసҠͷଛࣦܭࢉ த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝑚 1 𝑚 2 𝐿 2,1 -PTT GVOD 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 L2,1 (p2 , p1 ) #BDLQSPQ %FUBDI #BDLXBSE

Slide 21

Slide 21 text

w ϊʔυ̎ TPVSDF ͔Βϊʔυ̍ EFTUJOBUJPO ΁ͷ஌ࣝసҠ Τοδͷ஌ࣝసҠͷଛࣦܭࢉ த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝑚 1 𝑚 2 𝐿 2,1 -PTT GVOD 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 L2,1 (p2 , p1 ) Gate KL div 'PSXBSE p2 (c|x) p1 (c|x)

Slide 22

Slide 22 text

w ͲͷΑ͏ʹ஌ࣝసҠ͢Δ͔Λ੍ޚ ήʔτؔ਺ த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝑚 1 𝑚 2 𝐿 2,1 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 L2,1 (p2 , p1 ) 'PSXBSE p2 (c|x) p1 (c|x) Gate KL div $VUPGG(BUF -JOFBS(BUF 5ISPVHI(BUF $PSSFDU(BUF

Slide 23

Slide 23 text

w ೖྗ͞Εͨ஋Λͦͷ··ग़ྗ͢Δ ήʔτؔ਺ɿ5ISPVHI(BUF த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝑚 1 𝑚 2 𝐿 2,1 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 L2,1 (p2 , p1 ) $VUPGG(BUF -JOFBS(BUF $PSSFDU(BUF 5ISPVHI(BUF 𝐺 ( 𝐷 𝐾 𝐿 ) = 𝐷 𝐾 𝐿 มߋΛՃ͑ͣɺ ͦͷ··఻ୡ 'PSXBSE p2 (c|x) p1 (c|x) Gate KL div

Slide 24

Slide 24 text

w ೖྗʹରͯ͠ৗʹΛग़ྗ ήʔτؔ਺ɿ$VUP ff (BUF த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝑚 1 𝑚 2 𝐿 2,1 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 L2,1 (p2 , p1 ) $VUPGG(BUF -JOFBS(BUF $PSSFDU(BUF 5ISPVHI(BUF ৗʹΛग़ྗ Τοδͷ੾அ 𝐺 ( 𝐷 𝐾 𝐿 ) = 0 'PSXBSE p2 (c|x) p1 (c|x) Gate KL div

Slide 25

Slide 25 text

w ֶश͕࣌ؒܦաͱͱ΋ʹग़ྗ஋͕ঃʑʹେ͖͘ͳΔ ήʔτؔ਺ɿ-JOFBS(BUF த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝑚 1 𝑚 2 𝐿 2,1 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 L2,1 (p2 , p1 ) $VUPGG(BUF -JOFBS(BUF $PSSFDU(BUF 5ISPVHI(BUF Gate KL div ࣌ؒͱڞʹग़ྗ͕ େ͖͘ͳΔ 𝐺 ( 𝐷 𝐾 𝐿 ) = 𝑡 𝑡 𝑚 𝑎 𝑥 ∙ 𝐷 𝐾 𝐿 'PSXBSE p2 (c|x) p1 (c|x)

Slide 26

Slide 26 text

w ιʔεϊʔυ͕ਖ਼ղͨ͠৔߹ͷΈग़ྗ ήʔτؔ਺ɿ$PSSFDUHBUF த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝑚 1 𝑚 2 𝐿 2,1 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 L2,1 (p2 , p1 ) $VUPGG(BUF -JOFBS(BUF $PSSFDU(BUF 5ISPVHI(BUF Gate KL div ਖ਼ղͨ͠αϯϓϧͷ ৘ใͷΈ఻ୡ 𝐺 ( 𝐷 𝐾 𝐿 ) = 𝛿 ^ 𝑦 , 𝑦 𝑚 2 ∙ 𝐷 𝐾 𝐿 'PSXBSE p2 (c|x) p1 (c|x)

Slide 27

Slide 27 text

w ͲͷΑ͏ʹ஌ࣝసҠ͢Δ͔Λ੍ޚ ήʔτؔ਺ த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝑚 1 𝑚 2 𝐿 2,1 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 L2,1 (p2 , p1 ) 'PSXBSE p2 (c|x) p1 (c|x) Gate KL div $VUPGG(BUF -JOFBS(BUF 5ISPVHI(BUF $PSSFDU(BUF

Slide 28

Slide 28 text

w ϋΠύʔύϥϝʔλαʔνͰ஌ࣝసҠάϥϑΛ࠷దԽ ࠷దԽख๏"TZODISPOPVT4VDDFTTJWF)BMWJOH"MHPSJUIN "4)" ύϥϝʔλήʔτؔ਺ ิॿϊʔυ ஌ࣝసҠάϥϑͷ࠷దԽ த෦େֶϩΰ த෦େֶϩΰ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦 ήʔτؔ਺ • 5ISPVHI(BUF • $VUPGG(BUF • -JOFBS(BUF • $PSSFDU(BUF • 3FT/FU ධՁର৅ϊʔυ • 3FT/FU • 3FT/FU • 8JEF3FT/FU ิॿϊʔυ શ૊Έ߹Θͤ਺ɿ ௨Γʢϊʔυ਺ͷ৔߹ʣ

Slide 29

Slide 29 text

஌ࣝసҠάϥϑͷ࠷దԽ த෦େֶϩΰ த෦େֶϩΰ αʔό਺ɹɹɹɹɿ ୳ࡧճ਺ɹɹɹɹɿ ճ ϑϨʔϜϫʔΫɹɿ0QUVOB

Slide 30

Slide 30 text

w ධՁର৅ϊʔυɿ 3FT/FU w 7BOJMMBϞσϧɿ ࠷దԽʹΑͬͯ֫ಘͨ͠஌ࣝసҠάϥϑʢୈҐʣ த෦େֶϩΰ த෦େֶϩΰ ڭࢣϥϕϧ ධՁର৅ϊʔυ ิॿϊʔυ QSFUSBJOFE ิॿϊʔυ ڭࢣϥϕϧ

Slide 31

Slide 31 text

w ධՁର৅ϊʔυɿ 3FT/FU w 7BOJMMBϞσϧɿ ࠷దԽʹΑͬͯ֫ಘͨ͠஌ࣝసҠάϥϑʢୈҐʣ த෦େֶϩΰ த෦େֶϩΰ ิॿϊʔυ QSFUSBJOFE ิॿϊʔυ ධՁର৅ϊʔυ ஌ࣝৠཹ

Slide 32

Slide 32 text

w ධՁର৅ϊʔυɿ 3FT/FU w 7BOJMMBϞσϧɿ ࠷దԽʹΑͬͯ֫ಘͨ͠஌ࣝసҠάϥϑʢୈҐʣ த෦େֶϩΰ த෦େֶϩΰ ิॿϊʔυ QSFUSBJOFE ิॿϊʔυ ධՁର৅ϊʔυ ॳΊ͸,%ϥΠΫͳֶशɼ࣍ୈʹ,%ʴ%.-ͳֶश͕ߦΘΕΔ ɹ૬ޓֶशɹ

Slide 33

Slide 33 text

w ࣮ݧ֓ཁ σʔληοτɿ$*'"3 ֶशϊʔυ਺ɿ ࠷దԽର৅ϊʔυɿ3FT/FU ैདྷख๏ ,% %.- ͱͷൺֱ த෦େֶϩΰ த෦େֶϩΰ ख๏ ೝࣝ཰<> ิॿϊʔυͷϞσϧ *OEFQFOEFOU r ,% 3FT/FU QSFUSBJOFE %.- 3FT/FU 3FT/FU 0VST 3FT/FU QSFUSBJOFE 3FT/FU

Slide 34

Slide 34 text

࠷దԽʹΑͬͯ֫ಘͨ͠஌ࣝసҠάϥϑ $*'"3 த෦େֶϩΰ த෦େֶϩΰ ϊʔυ਺ɿ ϊʔυ਺ɿ ϊʔυ਺ɿ ϊʔυ਺ɿ ϊʔυ਺ɿ ϊʔυ਺ɿ *OEFQFOEFOU3FT/FU

Slide 35

Slide 35 text

࠷దԽʹΑͬͯ֫ಘͨ͠஌ࣝసҠάϥϑ $*'"3 த෦େֶϩΰ த෦େֶϩΰ *OEFQFOEFOU3FT/FU ϊʔυ਺ɿ

Slide 36

Slide 36 text

w ஌ࣝసҠάϥϑʹΞϯαϯϒϧϊʔυͱΞςϯγϣϯϩεΛಋೖ ΞςϯγϣϯϩεɿΤοδͷϩεʹΞςϯγϣϯϩεΛ௥Ճ Ξϯαϯϒϧϊʔυɿ֤ϊʔυͷग़ྗΛΞϯαϯϒϧ͢Δػߏ ஌ࣝసҠάϥϑΛ༻͍ͨΞϯαϯϒϧֶश<0LBNPUP &$$7> த෦େֶϩΰ த෦େֶϩΰ Ξϯαϯϒϧϊʔυ ^ 𝑦 𝐿 ^ 𝑦 , 𝑒 𝑒𝑛𝑠 𝑚 1 𝑚 2 𝑚 3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝐿 ^ 𝑦 ,3 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,1 𝐿 2,1 𝐿 1,2 𝐿 3,1 𝐿 1,3 𝐿 2,3 𝐿 3,2 𝐿 i,j = 𝐾 𝐿 ( 𝒑 i , 𝒑 j) ± 𝐿 𝐴 𝑇 ( 𝑸 i , 𝑸 j ) Ξςϯγϣϯϩε

Slide 37

Slide 37 text

w ϊʔυ͔̎Βϊʔυ̍΁ͷΞςϯγϣϯϩε ೋͭͷϞσϧؒͷΞςϯγϣϯΛ͚ۙͮͨΓ཭͢ޮՌ Ξςϯγϣϯϩε த෦େֶϩΰ த෦େֶϩΰ 4PVSDF %FTUJOBUJPO 𝑚 2 𝑚 1 p2 (c|x) p1 (c|x) 'PSXBSE 𝑚 1 𝑚 2 𝑚 3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝐿 ^ 𝑦 ,3 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,1 𝐿 2,1 𝐿 1,2 𝐿 3,1 𝐿 1,3 𝐿 2,3 𝐿 3,2 ,-EJW x x "UUFOUJPO "UUFOUJPO 𝑸 = 𝐶 ∑ 𝑖 =1 𝑨 𝑖 𝑝 ɿಛ௃Ϛοϓ ɿνϟωϧ਺ ɿϊϧϜ਺ 𝑨 𝑖 𝐶 𝑝 ैདྷͷଛࣦ 𝐿 𝐴𝑇 𝐿 𝐴𝑇 ( 𝑸 2 , 𝑸 1 ) = 1 𝐽 𝐽 ∑ 𝑗 𝑸 𝑗 2 𝑸 𝑗 2 2 − 𝑸 𝑗 1 𝑸 𝑗 1 2 2 "UUFOUJPOଛࣦ (BUF 𝐿 2,1 = G( 𝐾𝐿 ( 𝒑 2 , 𝒑 1) ± 𝐿 𝐴 𝑇 ( 𝑸 2 , 𝑸 1 )) ͚ۙͮΔ ཭͢ + −

Slide 38

Slide 38 text

w ,-EJWFSHFODFͱΞςϯγϣϯϩεΛ#BDLQSPQͯ͠ Λߋ৽ 𝑚 1 ϊʔυ̎ TPVSDF ͔Βϊʔυ̍ EFTUJOBUJPO ΁ͷ஌ࣝసҠ த෦େֶϩΰ த෦େֶϩΰ 𝑚 2 𝑚 1 𝑚 1 𝑚 2 𝑚 3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝐿 ^ 𝑦 ,3 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,1 𝐿 1,2 𝐿 3,1 𝐿 1,3 𝐿 2,3 𝐿 3,2 x x 4PVSDF %FTUJOBUJPO ͚ۙͮΔ ཭͢ + − 𝐿 2,1 #BDLXBSE 𝐿 2,1 = G( 𝐾𝐿 ( 𝒑 2 , 𝒑 1) ± 𝐿 𝐴 𝑇 ( 𝑸 2 , 𝑸 1 )) ,-EJW 𝐿 𝐴𝑇 (BUF "UUFOUJPO "UUFOUJPO Detach Back-prop

Slide 39

Slide 39 text

w ΞϯαϯϒϧϊʔυΛλʔήοτϊʔυͱͯ͠࠷େԽ͢ΔΑ͏ʹ࠷దԽ ֤ϊʔυͷग़ྗΛฏۉʹΑΓΞϯαϯϒϧ͢Δػߏ Ξϯαϯϒϧϊʔυ த෦େֶϩΰ த෦େֶϩΰ ^ 𝑦 𝐿 ^ 𝑦 , 𝑒 𝑒𝑛𝑠 𝑚 1 𝑚 2 𝑚 3 ^ 𝑦 ^ 𝑦 ^ 𝑦 𝐿 ^ 𝑦 ,3 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,1 𝐿 2,1 𝐿 1,2 𝐿 3,1 𝐿 1,3 𝐿 2,3 𝐿 3,2 𝑚 1 𝑚 2 𝑒𝑛𝑠 Ξϯαϯϒϧϊʔυ 𝑚 3 Ξϯαϯϒϧػߏ 𝑝 ( 𝑐 𝑥 ) = 𝑝 1 ( 𝑐 𝑥 ) + 𝑝 2 ( 𝑐 𝑥 ) + 𝑝 3 ( 𝑐 | 𝑥 ) 𝑝 1 ( 𝑐 𝑥 ) 𝑝 2 ( 𝑐 𝑥 ) 𝑝 3 ( 𝑐 𝑥 ) 𝑝 ( 𝑐 𝑥 )

Slide 40

Slide 40 text

஌ࣝసҠάϥϑͷΞϯαϯϒϧޮՌ த෦େֶϩΰ த෦େֶϩΰ ˠ஌ࣝసҠάϥϑʹ͓͍ͯΞϯαϯϒϧਫ਼౓͕޲্

Slide 41

Slide 41 text

࠷దԽͨ͠Ξϯαϯϒϧ஌ࣝసҠάϥϑ ϊʔυ਺ɿʣ த෦େֶϩΰ த෦େֶϩΰ ˠҟͳΔΞςϯγϣϯϚοϓʢΞϯαϯϒϧʹదͨ͠ଟ༷ੑʣΛ֫ಘ w Ξϯαϯϒϧϊʔυɿ w 7BOJMMBϞσϧɿ ೖྗը૾

Slide 42

Slide 42 text

࠷దԽͨ͠Ξϯαϯϒϧ஌ࣝసҠάϥϑ ϊʔυ਺ɿʣ த෦େֶϩΰ த෦େֶϩΰ ʹ͚ۙͮΔ ʹ͚ۙͮΔ ͔Β཭͢ ˠҟͳΔΞςϯγϣϯϚοϓʢΞϯαϯϒϧʹదͨ͠ଟ༷ੑʣΛ֫ಘ ʹ͚ۙͮΔ ͔Β཭͢ ͓ޓ͍ʹ͚ۙͮΔ ʹ͚ۙͮΔ ʹ͚ۙͮΔ ʹ͚ۙͮΔ ೖྗը૾ w Ξϯαϯϒϧϊʔυɿ w 7BOJMMBϞσϧɿ

Slide 43

Slide 43 text

ଟ༷ͳΞϯαϯϒϧϞσϧ͔Βͷ஌ࣝৠཹ த෦େֶϩΰ த෦େֶϩΰ w ஌ࣝసҠάϥϑͷΞϯαϯϒϧΛڭࢣͱͯ͠஌ࣝৠཹ ڭࢣωοτϫʔΫɿ஌ࣝసҠάϥϑͰֶशͨ͠ෳ਺ͷ"#/ʢ3FT/FUʣ ੜెωοτϫʔΫɿ"#/ʢ3FT/FUʣ ೖྗը૾ 𝒙 4UVEFOU/FUXPSL ωοτϫʔΫ 𝑚 1 𝒍 1 ( 𝒙 ) 𝒍 3 ( 𝒙 ) ωοτϫʔΫ 𝑚 3 𝒍 𝑒 𝑛 𝑠 ( 𝒙 ) 𝒑 𝑠 ( 𝒙 ) ڭࢣϥϕϧ ^ 𝑦 ஌ࣝసҠ 𝒑 𝑒 𝑛 𝑠 ( 𝒙 ) Թ౓෇͖4PGUNBYؔ਺ 5FBDIFS/FUXPSL 𝑒 𝑛 𝑠 𝑚 1 𝑚 2 𝑚 3 ஌ࣝసҠάϥϑ

Slide 44

Slide 44 text

ଟ༷ͳΞϯαϯϒϧϞσϧ͔Βͷ஌ࣝৠཹ த෦େֶϩΰ த෦େֶϩΰ ஌ࣝసҠάϥϑʹΑΔΞϯαϯϒϧΛৠཹ͢Δ͜ͱͰ ಉ͡ύϥϝʔλ਺Ͱߴ͍ೝࣝੑೳΛൃش

Slide 45

Slide 45 text

஌ࣝసҠάϥϑʹΑΔڞಉֶश த෦େֶϩΰ த෦େֶϩΰ w ஌ࣝసҠάϥϑΛఏҊ ̐छྨͷ(BUFؔ਺ʹΑΓɺ஌ࣝసҠΛ੍ޚ͢Δ͜ͱͰଟ༷ͳڞಉֶश ϋΠύʔύϥϝʔλαʔνʹΑΔ࠷దͳ஌ࣝసҠάϥϑΛ୳ࡧ ஌ࣝసҠάϥϑʹΑΔΞϯαϯϒϧֶश w ൃݟͨ͜͠ͱʢϊʔυ਺ͷ৔߹ʣ ,%ͱ%.-͕༥߹ͨ͠஌ࣝసҠάϥϑ͸ैདྷ๏Λ௒͑Δਫ਼౓Λୡ੒ ,%ͱ%.-ͷ༥߹ͨ͠஌ࣝసҠάϥϑ 𝑚 3 𝑚 1 𝑚 2 𝐿 ^ 𝑦 ,1 𝐿 ^ 𝑦 ,2 𝐿 ^ 𝑦 ,3 𝐿 1,2 𝐿 1,3 𝐿 2,1 𝐿 3,1 𝐿 3,2 𝐿 2,3 ^ 𝑦 ^ 𝑦 ^ 𝑦

Slide 46

Slide 46 text

ૉਓൃ૝ݰਓ࣮ߦ΁ த෦େֶϩΰ த෦େֶϩΰ w ஌ࣝసҠάϥϑΛఏҊ஌ࣝసҠ グ ϥϑʹΑΔڞಉֶश<.JOBNJ .*36 "$$7> ݚڀऀ͸໰୊ઃఆͱղ͖ํΛݶఆͤ ず ɼ஌ࣝసҠΛදݱ͢ΔϑϨʔϜϫʔΫΛݚڀऀ が ઐ໳ੑΛൃ شͯ͠ઃܭ ࠷దԽ୳ࡧΛߦ͏͜ͱ で ৽ͨͳ஌ݟΛൃݟ 4PGUXBSF࣌୅ͷݚڀํ๏ ܭࢉػ が ๲େͳ パ ϥϝʔλۭ͔ؒΒ࢓༷Λຬͨ͢ プ ϩ グ ϥϜΛ୳ࡧ ʮૉਓൃ૝ݰਓ࣮ߦʯ ୳ࡧʹΑΓ֫ಘͨ͠৽ͨͳ஌ݟ が ࣍ͷ৽ͨͳݚڀͷ୺ॹͱͳΔ͜ͱΛظ଴ ݚڀۭؒ ͜Ε·Ͱͷݚڀ ݚ ڀ ͷ ෳ ࡶ ͞ ૉਓൃ૝ݰਓ࣮ߦ

Slide 47

Slide 47 text

47 த෦େֶϩΰ த෦େֶϩΰ ஌ࣝసҠάϥϑʹΑΔ࠷దͳ൒ڭࢣ͋Γਂ૚ڞಉֶशͷ୳ࡧ<ଜຊ +4"*> ʮૉਓൃ૝ݰਓ࣮ߦʯ ୳ࡧʹΑΓ֫ಘͨ͠৽ͨͳ஌ݟ が ࣍ͷ৽ͨͳݚڀͷ୺ॹͱͳΔ͜ͱΛظ଴

Slide 48

Slide 48 text

w ϥϕϧ͋Γσʔλͱϥϕϧͳ͠σʔλΛֶशʹར༻ Ξϊςʔγϣϯʢϥϕϧ෇͚ʣʹ͔͔Δίετ࡟ݮ ֶश༻σʔλͷ֬อ͕༰қ ൒ڭࢣ͋Γֶशʢ4FNJTVQFSWJTFEMFBSOJOH 44- ֶश༻σʔλʹର͢Δσʔλͷׂ߹

Slide 49

Slide 49 text

w Ұகੑਖ਼ଇԽʢ$POTJTUFODZSFHVMBSJ[BUJPOʣ ϥϕϧͳ͠σʔλʹઁಈΛ෇༩͠ɼͦͷը૾ʹର͢ΔҰகੑΛֶश ैདྷ๏ɿ NPEFM<-BJOF *$-3>ɼ.FBO5FBDIFS<5BSWBJOFO /FVS*14>ͳͲ w ٖࣅϥϕϦϯάʢ1TFVEPMBCFMJOHʣ ༧ଌ݁ՌΛPOFIPUԽٖͯ͠ࣅϥϕϧΛϥϕϧͳ͠σʔλʹ෇༩ ϥϕϧ͋Γσʔλͱٖࣅϥϕϧ͋Γσʔλͷࠞ߹ηοτΛ༻͍ͯڭࢣ͋Γֶश ैདྷ๏ɿ1TFVEP-BCFM<-FF *$.->ͳͲ Π ൒ڭࢣ͋Γֶशͷ୅දతͳํ๏ ϥϕϧͳ͠σʔλ ༧ଌ /FUXPSL ٖࣅϥϕϧ͋Γσʔλ ʢϥϕϧͳ͠σʔλʣ )BSEUBSHFU ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ 予測1 摂動を付与 ٖࣅϥϕϦϯά ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ 予測1 摂動を付与 Ұ؏ੑਖ਼ଇԽ ϥϕϧͳ͠σʔλ ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ 予測1 摂動を付与 ʴઁಈ ʴઁಈ ༧ଌ ༧ଌ ޡࠩ /FUXPSL

Slide 50

Slide 50 text

w ٖࣅϥϕϦϯάɿऑม׵࣌ͷ༧ଌ͕ᮢ஋Λ௒͑ͨ৔߹ͷΈٖࣅϥϕϧΛੜ੒ ऑม׵ɿࠨӈ൓స ฏߦҠಈ w Ұகੑଛࣦɿੜ੒ٖͨ͠ࣅϥϕϧͱڧม׵࣌ͷ༧ଌͷޡࠩ ڧม׵ɿෳ਺छͷը૾ม׵ʹΑΔڧ͍ઁಈʢ3BOE"VHNFOU<$VCVLF $713>ʣ Ұகੑਖ਼ଇԽͱٖࣅϥϕϦϯάͷϋΠϒϦουɿ'JY.BUDI<4PIO /FVS*14> Ұ؏ੑଛࣦ ༧ଌ /FUXPSL ڧม׵ ऑม׵ ڭࢣ͋Γଛࣦ ϥϕϧ ༧ଌ ٖࣅϥϕϦϯά ϥϕϧ͋Γσʔλ ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ Network 予測1 予測2 摂動を付与 ϥϕϧͳ͠σʔλ )BSEUBSHFU ˠਓ͕ઃܭͨ͠૊Έ߹ΘͤͰ͋ΔͨΊ࠷దͳֶश๏ͱ͸ݶΒͳ͍

Slide 51

Slide 51 text

w ਓखʹΑΒͳ͍৽͍͠൒ڭࢣ͋Γڞಉֶश๏ͷ֫ಘ w Ξϓϩʔν ֤ैདྷ๏ΛͦΕͧΕάϥϑͰ౷Ұతʹදݱ άϥϑදݱͷߏ੒ཁૉΛϥϯμϜʹ૊Έ߹Θͤͯߴਫ਼౓ͳֶश๏Λ୳ࡧ ຊݚڀͷ໨త ɾɾɾ άϥϑදݱ NPEFM Π .FBO5FBDIFS !! !" Parameter for Exponential Moving Average KL-div ɾɾɾ NPEFM Π .FBO5FBDIFS ैདྷ๏ Network !! " +$! Network !! +$! ′ &(!! , " + $! ′) BackProp Loss Graphical representation !! !! !! &(!! , " + $! ) KL-div KL-div Network EMA(%! ) ' +)! Exponential Moving Average Network %! +)! ′ +(EMA(%! ), ' + )! ) +(%! , ' + )! ′) Loss BackProp !! !" Parameter for Exponential Moving Average Graphical representation KL-div ྫɿ NPEFM.FBO5FBDIFS Π άϥϑߏ଄ͷ୳ࡧ !! KL-div !" Parameter for Exponential Moving Average KL-div ɾɾɾ

Slide 52

Slide 52 text

ɾɾɾ άϥϑදݱ NPEFM Π .FBO5FBDIFS !! !" Parameter for Exponential Moving Average KL-div ɾɾɾ NPEFM Π .FBO5FBDIFS ैདྷ๏ Network !! " +$! Network !! +$! ′ &(!! , " + $! ′) BackProp Loss Graphical representation !! !! !! &(!! , " + $! ) KL-div KL-div Network EMA(%! ) ' +)! Exponential Moving Average Network %! +)! ′ +(EMA(%! ), ' + )! ) +(%! , ' + )! ′) Loss BackProp !! !" Parameter for Exponential Moving Average Graphical representation KL-div ྫɿ NPEFM.FBO5FBDIFS Π άϥϑߏ଄ͷ୳ࡧ !! KL-div !" Parameter for Exponential Moving Average KL-div ɾɾɾ w ֤ैདྷ๏ΛͦΕͧΕάϥϑͰ౷Ұతʹදݱ ૊Έ߹Θ͕ͤ༰қʹͳΓϋΠύʔύϥϝʔλͷΑ͏ʹௐ੔Մೳ ϊʔυɿωοτϫʔΫ Τοδɿଛࣦܭࢉ ैདྷͷ൒ڭࢣ͋Γֶश๏ΛάϥϑͰදݱ

Slide 53

Slide 53 text

w ڭࢣ͋ΓଛࣦͱҰகੑଛࣦ͕খ͘͞ͳΔΑ͏ʹֶश ڭࢣ͋Γଛࣦɿϥϕϧ͋Γσʔλͷ༧ଌͱϥϕϧͷޡࠩ Ұ؏ੑଛࣦɹɿϥϕϧͳ͠σʔλʹҟͳΔઁಈΛ෇༩ͨ࣌͠ͷ༧ଌؒͷޡࠩ ઁಈɿ%SPQPVUɼը૾ม׵ Ұகਖ਼ଇԽͷ୅දతͳख๏ɿ NPEFM<-BJOF *$-3> Π Ұகੑଛࣦ ༧ଌ /FUXPSL ڭࢣ͋Γଛࣦ ϥϕϧ ༧ଌ ϥϕϧ͋Γσʔλ ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ Network 予測1 予測2 摂動を付与 ϥϕϧͳ͠σʔλ ʴઁಈ ʴઁಈ )BSEUBSHFU ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ 予測1 摂動を付与

Slide 54

Slide 54 text

w Ұகੑଛࣦɿ࢝఺ͷϊʔυͱऴ఺ͷϊʔυ͕ಉ͡ΤοδͰදݱ ,-EJWFSHFODFʢ,-EJWʣͰ༧ଌؒͷޡࠩΛܭࢉ NPEFMΛάϥϑͰදݱ Π KL(f(x), f(x′  )) = C ∑ i fi (x)log fi (x) fi (x′  ) NPEFM Π άϥϑදݱ ɿ֬཰෼෍ʢ༧ଌ֬཰ʣ ɿΫϥε਺ f(x), f(x′  ) C Network !! " +$! Network !! +$! ′ &(!! , " + $! ′) BackProp Loss Graphical representation !! !! !! &(!! , " + $! ) KL-div KL-div

Slide 55

Slide 55 text

w NPEFMʹࢦ਺Ҡಈฏۉʢ&."ʣωοτϫʔΫΛಋೖֶͯ͠श ҰகੑଛࣦɿωοτϫʔΫͱ&."ωοτϫʔΫʹ͓͚Δ༧ଌؒͷޡࠩ &."ωοτϫʔΫͷॏΈ͸ωοτϫʔΫͷॏΈΛՃࢉ͢Δ͜ͱͰߋ৽ ॏΈύϥϝʔλͷΞϯαϯϒϧʹΑΓߴ͍ੑೳΛൃش͠ɼֶशΛิॿ Π Ұகੑਖ਼ଇԽͷ୅දతͳख๏ɿ.FBO5FBDIFS<5BSWBJOFO /FVS*14> ڭࢣ͋Γଛࣦ ϥϕϧ ϥϕϧ͋Γσʔλ ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ Network 予測1 予測2 摂動を付与 )BSEUBSHFU Ұகੑଛࣦ ༧ଌ /FUXPSL ༧ଌ ϥϕϧͳ͠σʔλ ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ Network 予測1 摂動を付与 ڭࢣ͋Γଛࣦ ϥϕϧ ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ! " #8 9 & ' ( ) *+, -./ 卡䇦: 卡䇦; ㏗俍> 冊卥 ϥϕϧ͋Γ σʔλ POFIPU ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ㏗俍> 冊卥 ϥϕϧͳ͠ σʔλ /FUXPSL ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ! " #8 9 & ' ( ) *+, -./ 卡䇦: ㏗俍> 冊卥 ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ༧ଌ ༧ଌ Ұ؏ੑଛࣦ ڭࢣ͋Γଛࣦ ϥϕϧ ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ! " #8 9 & ' ( ) *+, -./ 卡䇦: 卡䇦; ㏗俍> 冊卥 ϥϕϧ͋Γ σʔλ POFIPU ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ! " #8 9 & ' ( ) *+, -./ 卡䇦: 卡䇦; ㏗俍> 冊卥 ϥϕϧͳ͠ σʔλ /FUXPSL ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ! " #8 9 & ' ( ) *+, -./ 卡䇦: 卡䇦; ㏗俍> 冊卥 ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ! " #8 9 & ' ( 卡䇦: ㏗俍> 冊卥 ! " #$ % & ' ( ) *+, -./ ㎯1㌪刷 卡䇦 6〱 ! " #8 9 & ' ( 卡䇦: ㏗俍> 冊卥 ༧ଌ ༧ଌ Ұ؏ੑଛࣦ ʴઁಈ ʴઁಈ /FUXPSL

Slide 56

Slide 56 text

w &."ωοτϫʔΫʹՃࢉ͢Δύϥϝʔλͷํ޲ΛΤοδͰදݱ &."ωοτϫʔΫɿύϥϝʔλ ͷࢦ਺Ҡಈฏۉ஋ Ͱߋ৽ θ1 EMA(θ1 ) .FBO5FBDIFSΛάϥϑͰදݱ EMA(θ1,t ) = αEMA(θ1,t−1 ) + (1 − α)θ1,t ɿϋΠύʔύϥϝʔλ ɿֶशεςοϓ α t .FBO5FBDIFS άϥϑදݱ Network EMA(%! ) ' +)! Exponential Moving Average Network %! +)! ′ +(EMA(%! ), ' + )! ′) +(%! , ' + )! ) Loss BackProp !! !" Parameter for Exponential Moving Average Graphical representation KL-div

Slide 57

Slide 57 text

w ϥϕϧͳ͠σʔλʹٖࣅϥϕϧΛ෇༩ֶͯ͠श ֶशং൫͸ϥϕϧ͋ΓσʔλͷΈΛ༻͍ͯڭࢣ͋Γֶश ༧ଌΛ΋ͱʹϥϕϧͳ͠σʔλʹٖࣅϥϕϧΛ෇༩ʢ·ͨ͸ߋ৽ʣ ϥϕϧ͋Γσʔλͱٖࣅϥϕϧ͋Γσʔλͷࠞ߹ηοτΛ༻͍ͯڭࢣ͋Γֶश ̎ͱ̏Λ܁Γฦ͢ ϥϕϧ ༧ଌ ޡࠩ /FUXPSL ٖࣅϥϕϧ ͋Γσʔλ ࠞ߹ηοτ ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ Network 予測1 摂動を付与 ϥϕϧ͋Γ σʔλ ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ Network 予測1 予測2 摂動を付与 ٖࣅϥϕϧ͋Γσʔλ )BSEUBSHFU ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ Network 予測1 予測2 摂動を付与 ϥϕϧͳ͠σʔλ ༧ଌ /FUXPSL ラベルありデータ Network 正解情報 予測 誤差 ラベルなしデータ Network 予測1 予測2 摂動を付与 ٖࣅϥϕϦϯάͷ୅දతͳख๏ɿ1TFVEP-BCFM<-FF *$.->

Slide 58

Slide 58 text

w ٖࣅϥϕϧʹର͢Δ༧ଌͷޡࠩΛٻΊΔ1TFVEP-PTTʹΑΓΤοδͰදݱ 1TFVEP-PTTɿٖࣅϥϕϧͱ༧ଌ֬཰ͷޡࠩ 1TFVEP-BCFMΛάϥϑͰදݱ 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 Lpse(x) = E xy0logf(✓1, x + ⇣0 1 ) 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 Lpse(x) = E xy0logf(✓1, x + ⇣0 1 ) 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 Lpse(x) = E xy0logf(✓1, x + ⇣0 1 ) 1TFVEP-BCFM άϥϑදݱ ωοτϫʔΫ͕̎ͭͷ৔߹ Network !! or !" Pseudo-Labeling one-hot Network !! " +$! +$! ′ &(!! , " + $! ) Loss PseudoLoss !! !! !" PseudoLoss Graphical representation or BackProp

Slide 59

Slide 59 text

w ධՁର৅ϊʔυͷਫ਼౓͕࠷େԽ͢ΔΑ͏ʹάϥϑߏ଄Λ࠷దԽ ิॿϊʔυɿධՁର৅ϊʔυͷֶशΛαϙʔτ Losses: ・KL-divergence ・PseudoLoss Gate functions: ・Through Gate ・Cutoff Gate ・Linear Gate ・Threshold Gate Explore space: Models: ・ResNet32 ・WideResNet28-2 ・WideResNet28-6 ・EMA model Edge Node Backprop Detach Gate Loss ⋅ ධՁର৅ϊʔυ ิॿϊʔυ άϥϑ࠷దԽʹΑΔ൒ڭࢣ͋Γਂ૚ڞಉֶश๏ͷ୳ࡧ

Slide 60

Slide 60 text

ධՁ݁Ռ ϥϕϧ͋Γσʔλ਺ BMM 4VQFSWJTFE 1TFVEP-BCFM .FBO5FBDIFS 'JY.BUDI 0VST ϊʔυ 0VST ϊʔυ NPEFM Π ਖ਼ղ཰<> ˠϥϕϧ͋Γσʔλ਺ʹΑͬͯ࠷దͳ൒ڭࢣ͋Γֶश๏͸ҟͳΔ ଟ͍৔߹ʢ ʣɿ1TFVEP-BCFM͕ߴਫ਼౓ গͳ͍৔߹ʢ ʙ ʣɿ NPEFM .FBO5FBDIFS͕ߴਫ਼౓ Π

Slide 61

Slide 61 text

ධՁ݁Ռ ϥϕϧ͋Γσʔλ਺ BMM 4VQFSWJTFE 1TFVEP-BCFM .FBO5FBDIFS 'JY.BUDI 0VST ϊʔυ 0VST ϊʔυ NPEFM Π ਖ਼ղ཰<> ˠैདྷ๏ͱൺ΂୳ࡧͨ͠ख๏ʢ0VSTʣͷਫ਼౓͕࠷΋ߴ͍ ʴ̒ ʴ

Slide 62

Slide 62 text

w ࠷దԽͨ͠άϥϑߏ଄Λௐࠪ ൒ڭࢣ͋Γֶशͷ܏޲ w ܏޲ௐࠪ ֶशܦաʹ͓͚Δ൒ڭࢣ͋Γڞಉֶशͷ܏޲ ϥϕϧ͋Γσʔλ਺ʹ͓͚Δ൒ڭࢣ͋Γڞಉֶशͷ܏޲ ϊʔυ਺ͷมԽʹ͓͚Δ൒ڭࢣ͋Γڞಉֶशͷ܏޲ ࠷దԽͨ͠άϥϑߏ଄

Slide 63

Slide 63 text

w ϊʔυ਺ɿ̎ɼϥϕϧ͋Γσʔλ਺ɿ ʢগͳ͍ʣ ਖ਼ղ཰ɿ<> ࠷దԽͨ͠άϥϑߏ଄ɿֶशܦաʹ͓͚Δ܏޲<> (BUFؔ਺ ֶशํ๏ 1. ResNet32 (55.94%) Linear 2. WRN28_6 (57.81%) Linear Linear Label Through Label Through 1. ResNet32 (55.94%) Consistency 2. WRN28_6 (57.81%) KD Consistency Label Supervised Label Supervised ڭࢣ͋Γֶश ڭࢣ͋Γֶश ֶशং൫ɿݸʑͷϊʔυͰಠཱʹڭࢣ͋Γֶश

Slide 64

Slide 64 text

w ϊʔυ਺ɿ̎ɼϥϕϧ͋Γσʔλ਺ɿ ʢগͳ͍ʣ ਖ਼ղ཰ɿ<> ࠷దԽͨ͠άϥϑߏ଄ɿֶशܦաʹ͓͚Δ܏޲<> (BUFؔ਺ ֶशํ๏ 1. ResNet32 (55.94%) Linear 2. WRN28_6 (57.81%) Linear Linear Label Through Label Through ֶशং൫ɿݸʑͷϊʔυͰಠཱʹڭࢣ͋Γֶश 1. ResNet32 (55.94%) Consistency 2. WRN28_6 (57.81%) KD Consistency Label Supervised Label Supervised ˠֶशલ൒ͱޙ൒ͰҟͳΔֶशઓུ͕ޮՌత ֶशऴ൫ɿݸʑͷϞσϧ͕ NPEFMʹΑΔֶशʹมԽͭͭ͠ɼ̎൪ϊʔυ͔Βৠཹ Π NPEFM Π NPEFM Π ৠཹ

Slide 65

Slide 65 text

w ϊʔυ਺ɿ̎ɼϥϕϧ͋Γσʔλ਺ɿ ʢଟ͍ʣ ਖ਼ղ཰ɿ<> ࠷దԽͨ͠άϥϑߏ଄ɿֶशܦաʹ͓͚Δ܏޲<> (BUFؔ਺ ֶशํ๏ 1. ResNet32 (62.76%) Linear 2. WRN28_6 (60.4%) Through 1. ResNet32 (62.76%) Consistency 2. WRN28_6 (60.4%) PseudoLabeling ֶशࡁΈϞσϧ ٖࣅϥϕϦϯά ֶशং൫ɿࣄલֶशϞσϧͷٖࣅϥϕϦϯάͰֶश NPEFM Π ֶशऴ൫ɿٖࣅϥϕϦϯάͱ NPEFMͰֶश Π ˠֶशऴ൫ʹҰகੑਖ਼ଇԽʢ NPEFMʣͰֶश͢Δ͜ͱ͕ޮՌత Π

Slide 66

Slide 66 text

ֶशܦաʹ͓͚Δ܏޲ɿ6."1ʹΑΔՄࢹԽ FQPDI FQPDI 1. ResNet32 (62.76%) Consistency 2. WRN28_6 (60.4%) PseudoLabeling ֶशࡁΈϞσϧ ٖࣅϥϕϦϯά NPEFM Π ˠֶशޙ൒ͷҰகੑਖ਼ଇԽʢ NPEFMʣʹΑΓΫϥελԽ͕ਐΉ Π 1. ResNet32 (62.76%) Consistency 2. WRN28_6 (60.4%) PseudoLabeling ֶशࡁΈϞσϧ ٖࣅϥϕϦϯά

Slide 67

Slide 67 text

w ϥϕϧ͋Γσʔλ͕গͳ͍ʢ ຕʣ৔߹ ࠷దԽͨ͠άϥϑߏ଄ɿϥϕϧ͋Γσʔλ਺ʹ͓͚Δ܏޲<> 4,000 label 2,000 label 6,000 label 1. ResNet32 (55.94%) Consistency 2. WRN28_6 (57.81%) KD Consistency Label Supervised Label Supervised <> <> ֶशํ๏ʢϊʔυ਺̎ʣ ֶशํ๏ʢϊʔυ਺̏ʣ ϥϕϧ͋Γσʔλ͕গͳ͍৔߹ɿ Ұகੑਖ਼ଇԽʢ NPEFMʣͱৠཹʢ૬ޓֶशʣ͕ޮՌత Π ৠཹ NPEFM Π NPEFM Π NPEFM Π .FBO5FBDIFS ૬ޓֶश

Slide 68

Slide 68 text

w ϥϕϧ͋Γσʔλ͕ଟ͍ʢ ຕʣ৔߹ ࠷దԽͨ͠άϥϑߏ଄ɿϥϕϧ͋Γσʔλ਺ʹ͓͚Δ܏޲<> ֶशํ๏ʢϊʔυ਺̎ʣ ֶशํ๏ʢϊʔυ਺̏ʣ <> <> Gate関数 学習方法 Gate function Learning method 1. ResNet32 (62.76%) Consistency 2. WRN28_6 (60.4%) PseudoLabeling ֶशࡁΈϞσϧ ٖࣅϥϕϦϯά ٖࣅϥϕϦϯά ϥϕϧ͋Γσʔλ͕ଟ͍৔߹ɿ ˠධՁର৅ϊʔυ͸ٖࣅϥϕϦϯάΛ༻ֶ͍ͯश

Slide 69

Slide 69 text

w ϊʔυ਺ͷ৔߹ ࠷దԽͨ͠άϥϑߏ଄ɿϊʔυ਺͕ଟ͍࣌ͷ܏޲ ֶशํ๏ ʢϥϕϧ͋Γσʔλ ຕʣ ֶशํ๏ ʢϥϕϧ͋Γσʔλ ຕʣ <> <> Gate関数 学習方法 Gate function Learning method label 6,000 label 10,000 label 8,000 label .FBO5FBDIFS .FBO5FBDIFS ૬ޓֶश ิॿϊʔυ͸.FBO5FBDIFSΛ಺แˠࢦ਺ҠಈฏۉϞσϧͰΑΓྑ͍5FBDIFSΛֶश ϊʔυ਺͕ଟ͍৔߹͸.FBO5FBDIFSͰิॿϊʔυΛվળ͢Δ͜ͱ͕ޮՌత ɹˠ൒ڭࢣ͋Γڞಉֶश

Slide 70

Slide 70 text

w ධՁର৅ϊʔυʹରͯ͠खಈͰΤοδΛ௥Ճ ϥϕϧ͔ΒͷΤοδɹɹɹɹɹɹɿڭࢣ͋Γֶश ධՁର৅ϊʔυ͔Β̎΁ͷΤοδɿ.FBO5FBDIFSͱͷ૬ޓֶश ୳ࡧͰಘͨ஌ݟΛ׆͔ͨ͠खಈઃܭʹΑΔߋͳΔվળ ख࡞ۀͰઃܭͨ͠άϥϑʢʣ ୳ࡧͰ֫ಘͨ͠άϥϑʢʣ QU޲্ ୳ࡧͰಘͨάϥϑͱ஌ݟΛ׆͔͠खಈઃܭͰ͞Βʹվળ .FBO5FBDIFSͰิॿϊʔυΛվળͭͭ͠ ٖࣅϥϕϦϯάͰֶश .FBO5FBDIFSͰิॿϊʔυΛվળͭͭ͠ ٖࣅϥϕϦϯάͰ NPEFMͱ૬ޓֶश Π ֶशํ๏ ֶशํ๏ 学習方法 n Learning method .FBO5FBDIFS ٖࣅϥϕϦϯά Gate関数 学習方法 Gate function Learning method .FBO5FBDIFS ٖࣅϥϕϦϯά Ұ؏ੑਖ਼ଇԽ ૬ޓֶश 'FFECBDL ڭࢣ͋Γֶश -BCFM

Slide 71

Slide 71 text

w άϥϑ୳ࡧʹΑΓ৽͍͠൒ڭࢣ͋Γڞಉֶश๏Λ୳ࡧ άϥϑ୳ࡧʹΑΓಘΒΕͨޮՌతͳ൒ڭࢣ͋Γڞಉֶशʹ͓͚Δ஌ݟ ֶशͷܦաͱͱ΋ʹֶशઓུΛมԽͤ͞Δ͜ͱͰߴਫ਼౓Խ ֶशޙ൒ʹҰ؏ੑਖ਼ଇԽʢ NPEFMʣΛߦ͏͜ͱ͕༗ޮ ϥϕϧ͋Γσʔλ਺͝ͱʹ࠷దͳֶशઓུ͸ҟͳΔ ϥϕϧ͋Γσʔλ͕ଟ͍৔߹͸ٖࣅϥϕϦϯάͰͷֶश͕༗ޮɹ ෳ਺ϞσϧΛ༻͍ͨڞಉֶश͸൒ڭࢣ͋Γֶशʹ΋༗ޮ ϊʔυ਺͕ଟ͍৔߹͸.FBO5FBDIFSͰิॿϊʔυΛվળ͢Δ͜ͱͰޮՌతͳ൒ڭࢣ͋ΓڞಉֶशΛ࣮ݱ ୳ࡧͰಘͨάϥϑͱ஌ݟΛ׆͔͠खಈઃܭͰ͞Βʹվળ ϊʔυ਺ɼϥϕϧ͋Γσʔλ਺ ͷ୳ࡧͰಘͨάϥϑͷਫ਼౓ΛखಈઃܭͰQUվળ Π ·ͱΊɿ஌ࣝసҠάϥϑʹΑΔ࠷దͳ൒ڭࢣ͋Γਂ૚ڞಉֶशͷ୳ࡧ

Slide 72

Slide 72 text

ૉਓൃ૝ݰਓ࣮ߦ த෦େֶϩΰ த෦େֶϩΰ ʲϑΣϩʔ͔Βͷϝοηʔδʳ ৘ใɾγεςϜιαΠΤςΟࢽ ୈ 26 רୈ 4 ߸ʢ௨ר 105 ߸ʣ ૉਓൃ૝ݰਓ࣮ߦ 2.0 ϑΣϩʔ ౻٢ ߂࿱ த෦େֶ ʮண؟େہணखখہɼૉਓൃ૝ݰਓ࣮ߦʯ͸ɼ 2006 ೥ 8 ݄ʹࡏ֎ݚڀͰถࠃΧʔωΪʔϝϩϯ େֶϩϘοτ޻ֶݚڀॴʹ 1 ೥ؒ଺ࡏ͠ɼؼࠃ ͷࡍʹۚग़෢༤ઌੜ͔Β௖͍ͨݴ༿Ͱ͋Δɽ ʮૉ ਓൃ૝ݰਓ࣮ߦʯͱ͸ɼ ۚग़ઌੜͷஶॻ [1] ʹΑ Δͱɼ ʮൃ૝͸୯७ɼૉ௚ɼࣗ༝ɼ؆୯Ͱͳ͚Ε ͹ͳΒͳ͍ɽ͔͠͠ɼൃ૝Λ࣮ߦʹҠ͢ʹ͸஌ ͕͍ࣝΔɼख़࿅͞Εٕ͕͍ͨΔʯͱ͍͏͜ͱͰ ͋Δɽචऀ͸ͦΕҎདྷɼ͜ͷݴ༿ΛϞοτʔʹ ͯ͠ݚڀʹऔΓ૊ΜͰ͍Δɽ ͔͠͠ɼ ʮݴ͏͸қ ͘ɼߦ͏͸೉͠ʯͷయܕͰ͋Γɼ࣮ફ͢Δͷ͸ ͳ͔ͳ͔೉͍͠ɽଟ͘ͷ࿦จΛಡΜͰ͍͘ͱ஌ ͕ࣝਂ·Γઐ໳ੑ͸ߴ͘ͳΔ͕ɼͦΕ͕োนͱ ͳͬͯɼຊ࣭Ͱ͸ͳ͘খ͞ͳ͜ͱʹண໨ͨ͠໰ ୊ઃఆΛߦ͍͕ͪͰ͋Δɽ·ͨɼຊ࣭Λଊ͑ͯ ΋޻෉Λ۪ͤͣ௚ʹ࣮૷͢Δͱ͏·͘ಈ͔ͳ͍ ͜ͱ͕͋Δɽ ຊߘͰ͸ɼ ໿ 10 ೥Λܦͯʮૉਓൃ ૝ݰਓ࣮ߦʯʹগ͚͚ͩۙͮͨ͠ͷͰ͸ͱࢥ͏ චऀΒͷݚڀ 2 ྫʹ͍ͭͯ঺հ͠ɼ࠷ۙɼࣗ෼ ͳΓʹࢥ͍ඳ͘ʮૉਓൃ૝ݰਓ࣮ߦʯͷΞοϓ σʔτΛڞ༗͍ͨ͠ɽ 2010 ೥ࠒɼը૾ؒͷରԠ఺ϚονϯάͷͨΊ ͷಛ௃఺ݕग़ɾهड़ͷݚڀ͕ଟ͘औΓ૊·Εͯ ͍ͨɽதͰ΋ɼࣹӨมԽΛ൐͏ը૾ؒͷରԠ఺ Ϛονϯά͸ɼΩʔϙΠϯτͷಛ௃Λදݱ͢Δ ΞϑΟϯྖҬΛٻΊΔඞཁ͕͋Γɼ೉͍͠໰୊ Ͱ͋ͬͨɽैདྷख๏Ͱ͸ɼΩʔϙΠϯτʹର͠ ͯҰͭͷΞϑΟϯྖҬ͔͠ਪఆ͠ͳ͍ͨΊɼը ૾ͷมܗ΍ΩʔϙΠϯτͷҐஔͣΕͷӨڹʹΑ ΓҟͳΔΞϑΟϯྖҬΛਪఆͯ͠͠·͏ͱ͍͏ ໰୊͕͋ͬͨɽ͜Ε͸ɼہॴత୳ࡧΛߦ͏͜ͱ ͕ݪҼͰ͋Γɼ ʮண؟খہணखখہʯͱݴ͑Δɽ 2015 ೥ʹචऀΒ͕ࠃࡍձٞ ICCV ʹͯൃදͨ͠ ʮඇ౳ํੑ LoG ϑΟϧλʹΑΔෳ਺ͷΞϑΟϯ ྖҬͷਪఆʯ [2] Ͱ͸ɼ༷ʑͳପԁܗঢ়ͷඇ౳ํ ੑ LoG ϑΟϧλΛ༻͍ͯෳ਺ͷΞϑΟϯྖҬΛ ਪఆ͢Δ͜ͱΛఏҊͨ͠ɽγϯϓϧʹɼҰͭͰ ͸ͳ͘ෳ਺ͷྖҬ͕͋ͬͯ΋Α͍ͷͰ͸ɼͱ͍ ͏ʮૉਓൃ૝ʯͰ͋Δɽ ͔͠͠ɼ ͍࣮͟૷ͱͳΔ ͱɼ ඇ౳ํੑ LoG ϑΟϧλʹ͸ x ํ޲ͷεέʔ ϧɼy ํ޲ͷεέʔϧɼճస֯ͷ 3 ύϥϝʔλ ͕͋Γɼͦͷ૊Έ߹Θͤ͸਺ઍछྨͱͳΔɽෳ ਺ͷΞϑΟϯྖҬΛਪఆ͢ΔͨΊɼযΔ͕༨Γ ͜ͷ਺ઍछྨͷϑΟϧλશͯΛ৞ΈࠐΉॲཧΛ ͜ͷ··ߦ͏ͱɼ๲େͳܭࢉίετ͕ඞཁͱͳ Δɽ ͦ͜Ͱɼ ʮݰਓ࣮ߦʯͱͯ͠ɼ਺ઍछྨͷඇ ౳ํੑ LoG ϑΟϧλ܈Λಛҟ஋෼ղʹΑΓٻ Ίͨ 14 छྨͷݻ༗ϑΟϧλͰۙࣅ͠ɼ৞Έࠐ ΈॲཧΛޮ཰తʹܭࢉ͢Δ͜ͱʹͨ͠ɽ͜Εʹ ΑΓɼෳ਺ͷΞϑΟϯྖҬΛޮ཰తʹٻΊΔ͜ ͱ͕Ͱ͖ɼࣹӨมԽΛ൐͏ը૾ؒͷରԠ఺Ϛο νϯάͷߴਫ਼౓ԽΛ࣮ݱͨ͠ɽ͜ͷݚڀʹ͓͍ ͯɼ ʮૉਓൃ૝ݰਓ࣮ߦʯͷݴ༿͕ݚڀͷํ޲ੑ ΍ਐΊํΛܾΊΔखॿ͚Λͯ͘͠ΕͨΑ͏ʹࢥ ͑ɼ2006 ೥͔Β໿ 10 ೥ΛܦͯɼΑ͏΍͘ʮૉ ਓൃ૝ݰਓ࣮ߦʯʹҰา͚ۙͮͨͱࢥ͑Δݚڀ Ͱ͋ͬͨɽ͜ͷݚڀҎޙ΋ɼৗʹɼૉਓൃ૝Ͱ ݰਓ࣮ߦʹͳ͍ͬͯΔ͔Λࣗ໰ࣗ౴͠ͳ͕Βݚ ڀʹऔΓ૊ΜͰདྷͨɽ 2012 ೥Ҏ߱ɼਂ૚ֶश͕ओମͱͳͬͨίϯϐ 16 IUUQTXXXKTUBHFKTUHPKQBSUJDMFJFJDFJTTKPVSOBM@@BSUJDMFDIBSKB

Slide 73

Slide 73 text

ίϯϐϡʔλϏδϣϯ࠷લઢ த෦େֶϩΰ த෦େֶϩΰ IUUQTXXXLZPSJUTVQVCDPKQCPPLEFUBJM ίϯϐϡʔλϏδϣϯ࠷લઢ Spring 2022 ʗר಄ݴ 5 ר಄ݴ Spring 2022 Visual HullతνϡʔτϦΞϧͷεεϝ ˙౻٢߂࿱ 2000 ೥લޙʹऔΓ૊·Ε͍ͯͨίϯϐϡʔλϏδϣϯͷΞϧΰϦζϜͰ͋Δ ࢹମੵަࠩ๏ʢvisual hullʣΛ͝ଘ஌ͩΖ͏͔ɻࢹମੵަࠩ๏͸ɼLaurentini ͕ఏҊͨ͠ Shape-from-silhouette ʹΑΔ 3D ࠶ߏ੒ख๏Ͱ͋Δɻ·ͣɼΧϝ ϥࢹ఺͔ΒγϧΤοτը૾1) Λ༻͍ͯ෮ݩର৅ͷΦϒδΣΫτΛ౤Өͨ͠γϧ 1) γϧΤοτը૾ͷલܠϚεΫ ͸ɼ෮ݩରԠͰ͋ΔΦϒδΣ Ϋτͷ 2 ࣍ݩ౤ӨͰ͋Δɻ Τοτԁਲ਼Λ࡞੒͢ΔɻҟͳΔࢹ఺ͰࡱӨͨ͠γϧΤοτը૾͔Βੜ੒͞Εͨ ԁਲ਼ͷަ఺͸ Visual Hull ͱݺ͹Εɼ͜ͷަ఺ΛٻΊΔ͜ͱͰΦϒδΣΫτͷ 3 ࣍ݩܗঢ়ͷ෮ݩ͕ՄೳͱͳΔɻਤ 1 ͸ɼCV ͳΒͼʹ CG քͰ༗໊ͳ “Stanford Bunny” ʢhttp://graphics.stanford.edu/data/3Dscanrep/ʣ ͱݺ͹ΕΔ 3D Φ ϒδΣΫτΛ෮ݩͨ͠ྫͰ͋Δɻগͳ͍ࢹ఺ͷγϧΤοτը૾͔Β෮ݩͨ͠ 3 ࣍ݩܗঢ়͸ɼຊདྷͷόχʔͷ 3 ࣍ݩܗঢ়ʹͳ͍ͬͯͳ͍ɻҰํͰɼଟ਺ͷҟͳΔ ࢹ఺ͷγϧΤοτը૾Λ༻͍Δͱɼਖ਼֬ͳ 3 ࣍ݩܗঢ়Λ෮ݩ͢Δ͜ͱ͕Ͱ͖Δɻ ͜Ε͸ɼݪஶ࿦จΛಡΉ͜ͱʹ͓͍ͯ΋ಉ༷Ͱ͋Δͱࢲ͸ࢥ͏ɻ࿦จͷຊ࣭ ͕Ͳ͜ʹ͋Δ͔Λਂ͘ཧղ͢Δʹ͸ɼҰࢹ఺͔ΒಡΈࠐΉͷͰ͸ͳ͘ɼҟͳΔ (a) 3 ࢹ఺ (b) 80 ࢹ఺ ʜ A B C A B C γϧΤοτը૾ ෮ݩ݁Ռ ਤ 1 ࢹମੵަࠩ๏ʢvisual hullʣ ɻhttp://www.sanko-shoko.net/note.php? id=tjly ͷίʔυΛར༻ͯ͠࡞੒ɻ ί ϯϐϡʔλ Ϗδϣ ϯ࠷લઢɹ 4QSJOH Ҫ৲ળٱɾ ڇٱ঵޹ɾ ยԬ༟༤ɾ ౻٢߂࿱ฤ IUUQTXXXLZPSJUTVQVCDPKQCPPLEFUBJM

Slide 74

Slide 74 text

.13(›5PVS 74 த෦େֶϩΰ த෦େֶϩΰ IUUQTXXXZPVUVCFDPNXBUDI W(LV,'5&

Slide 75

Slide 75 text

ػց஌֮ϩϘςΟΫεݚڀάϧʔϓ த෦େֶϩΰ த෦େֶϩΰ ڭत ౻٢߂࿱ Hironobu Fujiyoshi E-mail: [email protected] 1997೥ த෦େֶେֶӃത࢜ޙظ՝ఔमྃ, 1997೥ ถΧʔωΪʔϝϩϯେֶϩϘοτ޻ֶݚڀॴPostdoctoral Fellow, 2000೥ த෦େֶ޻ֶ෦৘ใ޻ֶՊߨࢣ, 2004೥ த෦େֶ।ڭत, 2005೥ ถΧʔωΪʔϝϩϯେֶϩϘοτ޻ֶݚڀॴ٬һݚڀһ(ʙ2006೥), 2010೥ த෦େֶڭत, 2014೥໊ݹ԰େֶ٬һڭत.
 ܭࢉػࢹ֮ɼಈը૾ॲཧɼύλʔϯೝࣝɾཧղͷݚڀʹैࣄɽ
 ϩϘΧοϓݚڀ৆(2005೥)ɼ৘ใॲཧֶձ࿦จࢽCVIM༏ल࿦จ৆(2009೥)ɼ৘ใॲཧֶձࢁԼه೦ݚڀ৆(2009೥)ɼը૾ηϯγϯάγϯϙδ΢Ϝ༏लֶज़৆(2010, 2013, 2014೥) ɼ ిࢠ৘ใ௨৴ֶձ ৘ใɾγεςϜιαΠΤςΟ࿦จ৆(2013೥)ଞ ڭत ࢁԼོٛ Takayoshi Yamashita E-mail:[email protected] 2002೥ ಸྑઌ୺Պֶٕज़େֶӃେֶത࢜લظ՝ఔमྃ, 2002೥ ΦϜϩϯגࣜձࣾೖࣾ, 2009೥ த෦େֶେֶӃത࢜ޙظ՝ఔमྃ(ࣾձਓυΫλʔ), 2014೥ த෦େֶߨࢣɼ2017 ೥ த෦େֶ।ڭतɼ2021೥ த෦େֶڭतɽ
 ਓͷཧղʹ޲͚ͨಈը૾ॲཧɼύλʔϯೝࣝɾػցֶशͷݚڀʹैࣄɽ
 ը૾ηϯγϯάγϯϙδ΢Ϝߴ໦৆(2009೥)ɼిࢠ৘ใ௨৴ֶձ ৘ใɾγεςϜιαΠΤςΟ࿦จ৆(2013೥)ɼిࢠ৘ใ௨৴ֶձPRMUݚڀձݚڀ঑ྭ৆(2013೥)ड৆ɽ ߨࢣ ฏ઒ཌྷ Tsubasa Hirakawa E-mail:[email protected] 2013೥ ޿ౡେֶେֶӃത࢜՝ఔલظऴྃɼ2014೥ ޿ౡେֶେֶӃത࢜՝ఔޙظೖֶɼ2017೥ த෦େֶݚڀһ (ʙ2019೥)ɼ2017೥ ޿ౡେֶେֶӃത࢜ޙظ՝ఔमྃɽ2019 ೥ த෦େֶಛ೚ॿڭɼ2021೥ த෦େֶߨࢣɽ2014೥ ಠཱߦ੓๏ਓ೔ຊֶज़ৼڵձಛผݚڀһDC1ɽ2014೥ ESIEE Paris٬һݚڀһ (ʙ2015೥)ɽ ίϯϐϡʔλϏδϣϯɼύλʔϯೝࣝɼҩ༻ը૾ॲཧͷݚڀʹैࣄ