Tweet
Tweet

Slide 1

Slide 1 text

AI࠷৽࿦จಡΈձ2022೥10݄ ᷂tech vein ழມ ॆԝ

Slide 2

Slide 2 text

ࣗݾ঺հ ழມ ॆԝ (͍ͷ·ͨ ΈͭͻΖ) גࣜձࣾ tech vein / DeepRad גࣜձࣾ ֤୅දऔక໾ ݉ σϕϩούʔ twitter: @ino2222

Slide 3

Slide 3 text

Facebook άϧʔϓͷ঺հ IUUQTXXXGBDFCPPLDPNHSPVQT

Slide 4

Slide 4 text

ΞδΣϯμ https://paperswithcode.com/ ʹΑΔɺ arxiv.org ͷաڈ1ϲ݄ؒͷ࿦จτοϓ10ຊΛ ঺հɻ

Slide 5

Slide 5 text

Papers with Code https://www.arxiv-sanity.com/top

Slide 6

Slide 6 text

໨࣍

Slide 7

Slide 7 text

Top10 1. Git Re-Basin: Merging Models modulo Permutation Symmetries 2. QuestSim: Human Motion Tracking from Sparse Sensors with Simulated Avatars 3. Understanding Diffusion Models: A Uni fi ed Perspective 4. Transformers are Sample Ef fi cient World Models 5. Brain Imaging Generation with Latent Diffusion Models 6. Learning with Differentiable Algorithms 7. Operationalizing Machine Learning: An Interview Study 8. Decoding speech from non-invasive brain recordings 9. Faithful Reasoning Using Large Language Models 10. AudioLM: a Language Modeling Approach to Audio Generation

Slide 8

Slide 8 text

1. Git Re-Basinɻฒ΂׵͑ରশੑʹΑΔϞσϧͷϚʔδ (ݪจ: Git Re-Basin: Merging Models modulo Permutation Symmetries) σΟʔϓϥʔχϯάͷ੒ޭ͸ɺ͋Δछͷڊେͳඇತ࠷దԽ໰୊Λൺֱత༰қʹղ͘͜ͱ͕Ͱ͖ΔΑ͏ʹ ͳ͓͔ͬͨ͛Ͱ͋Δɻඇತ࠷దԽ͸NPࠔ೉Ͱ͋Δʹ΋͔͔ΘΒͣɺ୯७ͳΞϧΰϦζϜʢ͠͹͠͹֬཰ తޯ഑߱Լͷมछʣ͸ɺ࣮ࡍʹେن໛ͳχϡʔϥϧωοτϫʔΫʹద߹͢Δࡍʹڻ͘΂͖༗ޮੑΛࣔ ͢ɻզʑ͸ɺχϡʔϥϧωοτϫʔΫͷଛࣦϥϯυεέʔϓ͸ɺӅΕϢχοτͷ͢΂ͯͷՄೳͳॱྻର শੑΛߟྀͨ͠ޙɺʢ΄΅ʣ୯ҰͷྲྀҬ(basin)ΛؚΉ͜ͱΛओு͢Δɻզʑ͸ɺ͋ΔϞσϧͷϢχο τΛฒ΂ସ͑ͯɺࢀরϞσϧͷϢχοτͱҰகͤ͞ΔͨΊͷ3ͭͷΞϧΰϦζϜΛ঺հ͢Δɻ͜ͷม׵ ʹΑΓɺࢀরϞσϧʹ͍ۙತຍ஍ʹҐஔ͢Δɺػೳతʹ౳ՁͳॏΈͷू߹͕ੜ੒͞ΕΔɻ࣮ݧతʹ͸ɺ CIFAR-10ͱCIFAR-100Ͱಠཱʹֶशͨ͠ResNetϞσϧؒͷθϩόϦΞઢܗϞʔυ઀ଓΛॳΊ࣮ͯূ͢ ΔͳͲɺ༷ʑͳϞσϧΞʔΩςΫνϟͱσʔληοτͰ୯ҰྲྀҬݱ৅Λ࣮ূ͠·ͨ͠ɻ͞Βʹɺ༷ʑͳ Ϟσϧ΍σʔληοτʹ͓͍ͯɺϞσϧ෯΍ֶश࣌ؒͱϞʔυ઀ଓੑͱͷؒʹڵຯਂ͍ݱ৅͕͋Δ͜ͱ Λ໌Β͔ʹ͠·ͨ͠ɻ࠷ޙʹɺ୯ҰྲྀҬཧ࿦ͷܽ఺ʹ͍ͭͯɺઢܗϞʔυ݁߹Ծઆͷ൓ྫΛؚΊͯٞ࿦ ͢Δɻ w ໨తɿϞσϧ࠷దԽɾ࿈߹ֶशɾΞϯαϯϒϧֶशͷͨΊͷجૅݚڀ w ੒Ռɿ$*'"3 Ͱݸผʹֶशͨͭ͠ͷ3FT/FUϞσϧΛθϩόϦΞઢܗϞʔυ઀ଓ͢Δ͜ ͱ͕Ͱ͖ͨ w ํ๏ɿϞσϧϢχοτͷม׵ΞϧΰϦζϜͷಋೖ w ݻ༗໊ɿ(JU3F#BTJO w ஶऀॴଐɿϫγϯτϯେֶίϯϐϡʔλʔαΠΤϯεɾΤϯδχΞϦϯάֶ෦ https://arxiv.org/abs/2209.04836v1

Slide 9

Slide 9 text

Loss Landscape →Lossؔ਺ͷՄࢹԽख๏ IUUQTBSYJWPSHBCT

Slide 10

Slide 10 text

ଛࣦϥϯυεέʔϓ͕Single Basin ʢ୯Ұͷ͘΅Έʣ • SGDͰ࠷దղʹͨͲΓண͖΍͍͢ɻ

Slide 11

Slide 11 text

Linear Mode Connectivity (ઢܗϞʔυ઀ଓ) Ҿ༻࿦จ: Linear Mode Connectivity and the Lottery Ticket Hypothesis IUUQTBSYJWPSHBCT

Slide 12

Slide 12 text

Lottery Ticket Hypothesis (͘͡Ҿ͖Ծઆ) • େن໛ͳχϡʔϥϧωοτϫʔΫ͕ߴੑೳΛୡ੒͠΍͢ ͍ͷ͸ɼ͘͡Λͨ͘͞ΜҾ͍͍ͯΔ͔ΒͰ͸ͳ͍͔આɻ • Ϟσϧ಺ʹ౰ͨΓ͘͡ܦ࿏͕͋Δ͔ΒߴੑೳʹͳΔɺͭ ·Γ౰ͨΓ͘͡Ҏ֎͸מΓࠐΜͰ΋݁ՌʹӨڹ͕ͳ͍͸ ͣͱ͍͏࿦ࢫɻ • ʮઢܗิؒ(linear interpolation)ʹΑΔSGDϊΠζʹର ͯ҆͠ఆ͔ʁʯͰמΓࠐΈޮ཰Λௐ΂Δ͜ͱ͕Ͱ͖Δɻ ҆ఆ͞͸ෆ҆ఆੑղੳ(࣍ϖʔδ)Ͱௐ΂Δɻ

Slide 13

Slide 13 text

Linear instability(ઢܗෆ҆ఆ౓) • ͋ΔϞσϧ͕SDGϊΠζ(ϥϯμϜֶश࣌ͷมԽ)ʹରͯ҆͠ఆੑ͔Λௐ΂ Δํ๏ɻ • ಛఆͷ࣌఺ͷॏΈʹ͍ͭͯෆ҆ఆ౓ΛଌΔʹ͸ɺݸผʹϥϯμϜͳσʔλ (αϯϓϧɾaugumentation)ͰεςοϓT·Ͱֶशͨ͠ॏΈW1 T ͱW2 T Λઢ ܗʹมԽͤͨ࣌͞ͷଛࣦؔ਺(Τϥʔؔ਺)ͷ࠷େ஋ΛInstabilityͱ͢Δɻ͞ Βʹ k εςοϓ·Ͱֶशͯ͠ಉ༷ʹෆ҆ఆ౓ΛଌΓɺෆ҆ఆ౓͕૿͑ͳ͚ Ε͹҆ఆͱ͢Δɻ

Slide 14

Slide 14 text

Linear mode Connectivity • ̎ͭͷωοτϫʔΫ͕ޡࠩোนͷߴ͞(ϞσϧAɾBͷॏΈ WaʙWbΛઢܗิؒ࣌ͨ࣌͠ͷଛࣦؔ਺ͷ࠷େ஋)͕΄΅ 0(2%ҎԼ)ʹͳΔ(θϩόϦΞͷ)ܦ࿏͕͋Δ৔߹ʹϞʔυ઀ଓ ͞Ε͍ͯΔͱ͍͏ɻઢܗϞʔυ઀ଓ͞Ε͍ͯΔ෦෼ʢʹๅ͘ ͡ܦ࿏ʣҎ֎ΛמΓࠐΜͰ΋ਫ਼౓Λҡ࣋Ͱ͖ΔͷͰɺϞσϧ αΠζΛѹॖͰ͖Δɻ

Slide 15

Slide 15 text

Ҿ༻࿦จͰ͸ɺઢܗิؒͨ࣌͠ ͷؒͷਫ਼౓͸௿͍ • ෳࡶͳ՝୊(ImageNet)ɾϞσϧ(ResNet)ͳ΄Ͳinstabilityͷน(όϦΞ)͸ߴ ΊʹͳΔ ɻ • όϦΞ͕௿͍ʹๅ͘͡ޮՌ͕ߴ͍(מΓࠐΈѹॖͰ͖Δ)

Slide 16

Slide 16 text

ࠓճͷ࿦จख๏Ͱ͸ɺઢܗิؒ࣌ͷਫ਼ ౓্͕͕͍ͬͯΔ(ଛࣦ͕ݮ͍ͬͯΔ) • αΠζѹॖޮՌ͕ߴ͘ͳΔ

Slide 17

Slide 17 text

޻෉: ϞσϧBͷॏΈΛϞσϧAͷॏΈʹҰக ͢ΔΑ͏ɺಛఆͷϧʔϧͰฒͼସ͍͑ͯΔ • Activation matching 
 ׆ੑ౓ʹΑΔҰக౓ • Weight matching 
 ॏΈʹΑΔҰக౓ • STE(Straight-Through Estimator) matching 
 ετϨʔτεϧʔਪఆྔʢSTEʣΛ࢖ͬͨҰக ౓

Slide 18

Slide 18 text

2. QuestSim: ໛ٖΞόλʔʹΑΔૄͳηϯαʔ͔Βͷਓମӡಈ௥੻ (ݪจ: QuestSim: Human Motion Tracking from Sparse Sensors with Simulated Avatars) AR/VRʹ͓͚ΔΠϯλϥΫςΟϒͰ຅ೖײͷ͋Δମݧʹ͸ɺਓମͷಈ͖ΛϦΞϧλΠϜʹτϥο Ωϯά͢Δ͜ͱ͕ॏཁͰ͋Δɻ͔͠͠ɺHMDʢHead Mounted Devicesʣ΍ARάϥεͳͲͷ୯ ମͷ΢ΣΞϥϒϧσόΠε͔ΒಘΒΕΔ਎ମʹؔ͢Δηϯασʔλ͸ඇৗʹݶΒΕͨ΋ͷͰ͋ ΔɻຊݚڀͰ͸ɺHMDͱ2ͭͷίϯτϩʔϥ͔Βͷૄͳ৴߸ΛऔΓࠐΈɺ΋ͬͱ΋Β͘͠ɺ෺ ཧతʹଥ౰ͳશ਎ӡಈΛγϛϡϨʔτ͢ΔڧԽֶशͷϑϨʔϜϫʔΫΛఏࣔ͢ΔɻຊݚڀͰ ͸ɺHMDͱ2ͭͷίϯτϩʔϥ͔Βͷૄͳ৴߸Λೖྗͱ͠ɺ෺ཧతʹଥ౰ͳશ਎ӡಈΛγϛϡ Ϩʔτ͢ΔڧԽֶशϑϨʔϜϫʔΫΛఏҊ͢Δɻͦͷ݁ՌɺHMDͷ6࣍ݩมܗͷΈΛೖྗͱ͠ ͨ৔߹Ͱ΋ɺԼ൒਎ΛҰ੾؍ଌ͢Δ͜ͱͳ͘ɺڻ͘΄Ͳਅ࣮ͱಉ͡Α͏ͳ٭ͷಈ͖Λ͢Δ͜ͱ Λ࣮ূͨ͠ɻ·ͨɺ୯ҰͷϙϦγʔͰɺଟ༷ͳӡಈελΠϧɺҟͳΔମ֨ɺ৽نͷ؀ڥʹର͠ ͯϩόετͰ͋Δ͜ͱΛࣔ͢ɻ w ໨తɿ"373ͷ຅ೖײվળ w ੒Ռɿ).%ʴ̎ίϯτϩʔϥ͚ͩͰશ਎ӡಈΛγϛϡϨʔτ͢ΔֶशϑϨʔϜϫʔΫΛఏҊ w ํ๏ɿ෺ཧۭؒ .FUB2VFTU ΍Ծ૝ۭؒ *TTBDHZN ͰͭͷσόΠεͷ࣠৴߸ೖྗΛ༻͍ ͯڧԽֶश w ݻ༗໊ɿ2VFTU4JN w ஶऀॴଐɿ.FUB https://arxiv.org/abs/2209.09391v1

Slide 19

Slide 19 text

HMD(ͱίϯτϩʔϥ)ͷ 
 3࣍ݩҐஔ(3࣠)+֯౓(3࣠)ͷ࠲ඪͱ଎౓͔Β࢟੎Λਪఆ • ࠨ: ̏σόΠε(಄+྆ख) / ӈ:̍σόΠε(಄)

Slide 20

Slide 20 text

No content

Slide 21

Slide 21 text

No content

Slide 22

Slide 22 text

ֶशσʔλɾֶश؀ڥ • σʔληοτ:ඃݧऀ172໊ͷ߹ܭ8࣌ؒͷϞʔ γϣϯΫϦοϓ(ө૾ͱMetaQuestͷ࠲ඪɾ଎ ౓৘ใ) • ֶश؀ڥ: Nvidia PhysXͱRLֶशϑϨʔϜϫʔ ΫIssac gym

Slide 23

Slide 23 text

಄ͷηϯαʔ͚ͩͰԼ൒਎ͷಈ ࡞ΛೝࣝͰ͖͍ͯΔ

Slide 24

Slide 24 text

YoutubeσϞϏσΦ https://www.youtube.com/watch?v=CkTHsz6Ldas

Slide 25

Slide 25 text

Limitations • ະֶशͷಈ࡞͸׬શʹτϥοΩϯάͰ͖ͳ͍ɻ ෳࡶͳಈ࡞͸ਫ਼౓͕མͪͨΓϒϨͨΓ͢Δɻ • ະֶशͷμΠφϛοΫͳಈ࡞ʢϒϨΠΫμϯεɾ δϟϯϓͳͲʣ͸స౗ͯ͠͠·͏͜ͱ͕͋Δɻ • ্൒਎ͱԼ൒਎͕૬ؔͷͳ͍ಈ͖Ͱ͸ɺϢʔβ ͷϙʔζͱγϛϡϨʔγϣϯ͕Ұக͠ͳ͍Մೳੑ ͕͋Δɻ

Slide 26

Slide 26 text

3. ֦ࢄϞσϧͷཧղɻ౷Ұతͳࢹ఺ (ݪจ: Understanding Diffusion Models: A Uni fi ed Perspective) ֦ࢄϞσϧ͸ੜ੒Ϟσϧͱͯ͠ڻ͘΂͖ೳྗΛ͓ࣔͯ͠Γɺ࣮ࡍɺImagen ΍ DALL-E 2 ͳͲͷςΩε τ৚݅෇͖ը૾ੜ੒ʹؔ͢Δݱࡏͷ࠷ઌ୺ϞσϧΛࢧ͍͑ͯΔɻ͜ͷݚڀͰ͸ɺม෼๏ͱείΞϕʔ εͷ྆ํͷ؍఺͔Β֦ࢄϞσϧͷཧղΛݟ௚͠ɺṖΛղ͖ɺ౷Ұ͢Δɻ·ͣɺϚϧίϑܕ֊૚తม෼ ΦʔτΤϯίʔμͷಛघͳέʔεͱͯ͠ม෼֦ࢄϞσϧʢVDMʣΛಋग़͠ɺ3ͭͷॏཁͳԾఆʹΑ ΓɺELBOͷѻ͍΍͍͢ܭࢉͱεέʔϥϒϧͳ࠷దԽΛՄೳʹ͠·͢ɻVDMΛ࠷దԽ͢Δ͜ͱ͸ɺ3ͭ ͷજࡏతͳ໨తͷ1ͭΛ༧ଌ͢ΔͨΊͷχϡʔϥϧωοτϫʔΫͷֶशʹؼண͢Δ͜ͱΛূ໌͠·͢ɻ ͦΕ͸ɺ೚ҙͷϊΠζԽ͞Εͨೖྗ͔ΒݩͷιʔεೖྗΛ༧ଌ͢Δ͜ͱɺ೚ҙͷϊΠζԽ͞Εͨೖྗ ͔ΒݩͷιʔεϊΠζΛ༧ଌ͢Δ͜ͱɺ೚ҙͷϊΠζϨϕϧʹ͓͍ͯϊΠζԽͨ͠ೖྗͷείΞؔ਺ Λ༧ଌ͢Δ͜ͱɺͰ͢ɻ࣍ʹɺείΞؔ਺Λֶश͢Δ͜ͱͷҙຯΛਂ͘۷ΓԼ͛ɺTweedieͷࣜΛ௨ ͯ͠ɺ֦ࢄϞσϧͷม෼తͳ؍఺ͱείΞϕʔεͷੜ੒ϞσϦϯάͷ؍఺Λ໌ࣔతʹ݁ͼ͚ͭΔɻ࠷ ޙʹɺ֦ࢄϞσϧΛ༻͍ͨ৚݅෇͖෼෍ͷֶशํ๏ʹ͍ͭͯɺΨΠμϯεΛ௨ͯ͡આ໌͢Δɻ w ໨తɿ%J ff VTJPO.PEFMΛ౷Ұղऍ͢Δ w ੒Ռɿ&-#0ͷѻ͍΍͍͢ܭࢉͱεέʔϥϒϧͳ࠷దԽΛՄೳʹͨ͠ w ํ๏ɿ w ݻ༗໊ɿ w ஶऀॴଐɿ(PPHMF3FTFBSDI #SBJO5FBN https://arxiv.org/abs/2208.11970v1

Slide 27

Slide 27 text

No content

Slide 28

Slide 28 text

ELBO; evidence lower bound • ม෼ਪ࿦Ϟσϧ͕؍ଌσʔλΛͲΕ͘Β͍આ ໌Ͱ͖͍ͯΔ͔ͷධՁࢦඪ

Slide 29

Slide 29 text

VAE ม෼ΦʔτΤϯίʔμ

Slide 30

Slide 30 text

HVAE ֊૚ܕม෼ΦʔτΤϯίʔμ • ଟ૚֦ுVAEΛҰൠԽͨ͠΋ͷ

Slide 31

Slide 31 text

VDM ม෼཭ࢄϞσϧ • ʹ੍ݶ͖ͭϚϧίϑܕHVAEʢMHVAEʣ

Slide 32

Slide 32 text

4. τϥϯεϑΥʔϚʔ͸αϯϓϧޮ཰తͳੈքϞσϧ (ݪจ: Transformers are Sample Ef fi cient World Models) ਂ૚ڧԽֶशΤʔδΣϯτ͸αϯϓϧޮ཰͕ѱ͍͜ͱͰ༗໊Ͱ͋Γɺ࣮໰୊΁ͷద༻͕͔ͳΓ੍ݶ͞ Ε͍ͯΔɻۙ೥ɺ͜ͷ໰୊Λղܾ͢ΔͨΊʹଟ͘ͷϞσϧϕʔεख๏͕ߟҊ͞ΕɺੈքϞσϧΛ૝૾ ͠ͳ͕Βֶश͢Δํ๏͕࠷΋ݦஶͳΞϓϩʔνͷҰͭͰ͋Δɻ͔͠͠ɺγϛϡϨʔγϣϯ؀ڥͱͷࣄ ্࣮ແ੍ݶͷΠϯλϥΫγϣϯ͸ັྗతʹฉ͑͜Δ͕ɺੈքϞσϧ͸௕࣌ؒʹΘͨͬͯਖ਼֬Ͱͳ͚Ε ͹ͳΒͳ͍ɻτϥϯεϑΥʔϚʔ͕γʔέϯεϞσϦϯά՝୊Ͱ੒ޭͨ͜͠ͱʹಈػ͚ͮΒΕɺզʑ ͸཭ࢄΦʔτΤϯίʔμͱࣗݾճؼτϥϯεϑΥʔϚʔ͔ΒͳΔੈքϞσϧͰֶश͢Δσʔλޮ཰ͷ ྑ͍ΤʔδΣϯτɺIRISΛ঺հ͢ΔɻΞλϦ100kϕϯνϚʔΫͰΘ͔ͣ2࣌ؒͷήʔϜϓϨΠʹ૬౰ ͢ΔIRIS͸ɺਓؒͷਖ਼نԽฏۉείΞ1.046Λୡ੒͠ɺ26ήʔϜத10ήʔϜͰਓؒΛ྇կ͢ΔੑೳΛ ൃش͢Δɻ͜ͷख๏͸ɺϧοΫϔου୳ࡧΛ༻͍ͳ͍ख๏ͱͯ͠৽ͨͳ஍ҐΛཱ֬͠ɺ͞ΒʹMuZero Λ྇կ͍ͯ͠·͢ɻαϯϓϧޮ཰ͷྑ͍ڧԽֶशͷͨΊͷTransformersͱੈքϞσϧʹؔ͢Δࠓޙͷ ݚڀΛଅਐ͢ΔͨΊɺࢲͨͪͷίʔυϕʔεΛhttps://github.com/eloialonso/iris Ͱެ։͠·͢ɻ w ໨తɿਂ૚ڧԽֶशͷֶशޮ཰ͷվળ w ੒Ռɿݱ࣮తͳֶश࣌ؒͰֶ΂Δσʔλޮ཰ͷྑ͍ΤʔδΣϯτΛ։ൃ w ํ๏ɿ཭ࢄΦʔτΤϯίʔμࣗݾճؼτϥϯεϑΥʔϚʔΛ࢖ͬͯɺ૝૾ੈքͰֶश͢Δ w ݻ༗໊ɿ*3*4 w ஶऀॴଐɿδϡωʔϰେֶ εΠε https://arxiv.org/abs/2209.00588v1

Slide 33

Slide 33 text

No content

Slide 34

Slide 34 text

ݱ࣮(྘)͔ΒੈքϞσϧ(G)Λ ࢖ͬͯγϛϡϨʔγϣϯֶश

Slide 35

Slide 35 text

ੈքϞσϧ(G)ͷ༧ଌྫ ্:࣮؀ڥ, Լ:྘࿮͔Βਪ࿦ͨ݁͠Ռ

Slide 36

Slide 36 text

ੈքϞσϧͷ඼࣭޲্͕؊ • ཭ࢄΦʔτΤϯίʔμͰɺήʔϜཁૉ(Ϙʔ ϧɾϓϨΠϠʔɾఢ)Λਖ਼͘͠࠶ߏங͢Δ • τϥϯεϑΥʔϚʔͰɺใुͷؼଐ΍Τϐ ιʔυͷऴྃΛଊ͑Δ

Slide 37

Slide 37 text

No content

Slide 38

Slide 38 text

5. જࡏత֦ࢄϞσϧʹΑΔ೴ը૾ੜ੒ (ݪจ: Brain Imaging Generation with Latent Diffusion Models) σΟʔϓχϡʔϥϧωοτϫʔΫ͸ɺҩྍը૾ղੳʹ໨֮·͍͠ϒϨʔΫεϧʔΛ΋ͨΒ͍ͯ͠ ·͢ɻ͔͠͠ɺͦͷσʔλϋϯάϦʔͳੑ࣭͔Βɺҩྍը૾ϓϩδΣΫτʹ͓͚Δ߇͑Ίͳσʔ ληοταΠζ͸ɺͦͷજࡏೳྗΛे෼ʹൃش͢Δ๦͛ʹͳ͍ͬͯΔՄೳੑ͕͋Γ·͢ɻ߹੒ σʔλͷੜ੒͸ɺֶशσʔληοτΛิ׬͠ɺΑΓେن໛ͳҩ༻ը૾ݚڀΛՄೳʹ͢Δ༗๬ͳ୅ ସखஈΛఏڙ͠·͢ɻ֦ࢄϞσϧ͸࠷ۙɺϑΥτϦΞϦεςΟοΫͳ߹੒ը૾Λੜ੒͢Δ͜ͱͰ ίϯϐϡʔλϏδϣϯͷίϛϡχςΟͷ஫໨ΛूΊ͍ͯΔɻຊݚڀͰ͸ɼߴղ૾౓೴ը૾͔Β߹ ੒ը૾Λੜ੒͢ΔͨΊʹɼજࡏత֦ࢄϞσϧΛར༻͢Δ͜ͱΛݕ౼͢ΔɽUK Biobankσʔλ ηοτʢN=31,740ʣͷT1w MRIը૾Λ༻͍ͯɺ೥ྸɺੑผɺ೴ߏ଄ମੵͳͲͷڞม਺Λ৚݅ ͱͯ͠ɺ೴ը૾ͷ֬཰త෼෍Λֶश͢ΔϞσϧΛߏஙͨ͠ɻͦͷ݁ՌɼզʑͷϞσϧ͸ݱ࣮తͳ σʔλΛੜ੒͢Δ͜ͱ͕Ͱ͖ɼ৚݅෇͚ม਺Λ༻͍ͯσʔλੜ੒ΛޮՌతʹ੍ޚͰ͖Δ͜ͱ͕Θ ͔ͬͨɽ·ͨɼ10ສຕͷ೴ը૾͔ΒͳΔ߹੒σʔληοτΛ࡞੒͠ɼՊֶքʹެ։ͨ͠ɽ w ໨తɿֶश༻ҩྍը૾σʔληοτͷෆ଍Λิ͏ w ੒Ռɿສຕͷ߹੒೴ը૾σʔληοτΛެ։ w ํ๏ɿ೴ը૾ͷ֬཰෼෍Λֶश͢ΔϞσϧ -%.-BUFOU%J ff VTJPO.PEFMT Λߏங w ݻ༗໊ɿ w ஶऀॴଐɿΩϯάεɾΧϨοδɾϩϯυϯΞϝϦΧࠃཱਫ਼ਆӴੜݚڀॴͳͲ https://arxiv.org/abs/2209.07162v1

Slide 39

Slide 39 text

No content

Slide 40

Slide 40 text

ੜ੒ը૾ͷఆྔධՁ

Slide 41

Slide 41 text

ڞมྔͰ৚݅෇͚ͯ͠ը૾ੜ੒ • ೥ྸ • ੑผ • ೴ߏ଄ମ༰ੵͳͲ • কདྷతʹ͸ɺը૾΍์ࣹઢϨϙʔτ΋ର৅ʹ ͢Δ༧ఆ

Slide 42

Slide 42 text

https://www.healthdatagateway.org/ (ͳͲ)Ͱ೴ը૾σʔληοτެ։

Slide 43

Slide 43 text

6. ඍ෼ՄೳͳΞϧΰϦζϜʹΑΔֶश (ݪจ: Learning with Differentiable Algorithms) ݹయతͳΞϧΰϦζϜͱχϡʔϥϧωοτϫʔΫͷΑ͏ͳػցֶशγεςϜ͸ɺͲͪΒ΋೔ৗੜ׆ʹᷓΕͯ ͍ΔɻݹయతͳίϯϐϡʔλαΠΤϯεͷΞϧΰϦζϜ͸ɺେ͖ͳάϥϑͷ࠷୹ܦ࿏ΛٻΊΔΑ͏ͳݫີʹ ఆٛ͞ΕͨλεΫͷਖ਼֬ͳ࣮ߦʹద͍ͯ͠Δ͕ɺχϡʔϥϧωοτϫʔΫ͸ɺਖ਼֬ͳΞϧΰϦζϜʹؐݩͰ ͖ͳ͍ը૾෼ྨͷΑ͏ͳΑΓෳࡶͳλεΫʹ͓͍ͯɺσʔλ͔Βֶशͯ͠࠷΋Մೳੑͷߴ͍౴͑Λ༧ଌ͢Δ ͜ͱ͕ՄೳͰ͋Δɻຊ࿦จͰ͸ɺ྆ऀͷ௕ॴΛੜ͔ͨ͢ΊʹɺΑΓؤ݈ͰɺΑΓߴੑೳͰɺΑΓղऍ͠΍͢ ͘ɺΑΓܭࢉޮ཰ͷߴ͍ɺΑΓσʔλޮ཰ͷߴ͍ΞʔΩςΫνϟʹͭͳ͕Δ྆֓೦ͷ݁߹Λ୳ٻ͍ͯ͠Δɻ ຊ࿦จͰ͸ɺχϡʔϥϧωοτ͕ΞϧΰϦζϜ͔Βɺ͋Δ͍͸ΞϧΰϦζϜͱ࿈ܞֶͯ͠श͢Δ͜ͱΛՄೳ ʹ͢ΔΞϧΰϦζϜ؂ࢹͱ͍͏ߟ͑ํΛఆࣜԽ͠·͢ɻΞϧΰϦζϜΛχϡʔϥϧɾΞʔΩςΫνϟʹ૊Έ ࠐΉ৔߹ɺΞʔΩςΫνϟΛΤϯυπʔΤϯυͰֶशͤ͞ɺޯ഑ΛΞϧΰϦζϜʹ༗ҙٛʹ఻ൖͤ͞Δ͜ͱ ͕Ͱ͖ΔΑ͏ɺΞϧΰϦζϜ͕ඍ෼ՄೳͰ͋Δ͜ͱ͕ॏཁͰ͋Δɻຊ࿦จͰ͸ɺΞϧΰϦζϜΛඍ෼Մೳʹ ͢ΔͨΊʹɺม਺ʹઁಈΛ༩͑ɺดͨ͡ܗͰɺ͢ͳΘͪαϯϓϦϯάͳ͠Ͱظ଴஋Λۙࣅ͢Δ͜ͱʹΑͬ ͯɺΞϧΰϦζϜΛ࿈ଓతʹ؇࿨͢ΔҰൠతͳํ๏ΛఏҊ͢Δɻ͞Βʹɺඍ෼ՄೳͳιʔτωοτϫʔΫɺ ඍ෼ՄೳͳϨϯμϥʔɺඍ෼Մೳͳ࿦ཧήʔτωοτϫʔΫͱ͍ͬͨඍ෼ՄೳͳΞϧΰϦζϜΛఏҊ͢Δɻ ࠷ޙʹɺຊ࿦จ͸ΞϧΰϦζϜΛ༻ֶ͍ͨशͷͨΊͷ୅ସతͳֶशઓུΛఏࣔ͢Δɻ w ໨తɿݹయΞϧΰϦζϜΛχϡʔϥϧωοτϫʔΫʹ׆͔͍ͨ͠ w ੒Ռɿඍ෼ՄೳͳΞϧΰϦζϜͷ։ൃͱϥΠϒϥϦఏڙ w ํ๏ɿΞϧΰϦζϜʹΑΔ؂ࢹ 4VQFSWJTPO ϑϨʔϜϫʔΫΛఆٛɾ։ൃ w ݻ༗໊ɿ w ஶऀॴଐɿίϯελϯπେֶ υΠπ https://arxiv.org/abs/2209.00616v1

Slide 44

Slide 44 text

ݹయΞϧΰϦζϜΛNNͷੈքʹ ૊ΈࠐΉ • ྫ: खॻ͖จࣈΛจࣈೝࣝͯ͠ιʔτ͢ΔϞσ ϧ

Slide 45

Slide 45 text

ίʔυྫ IUUQTHJUIVCDPN'FMJY1FUFSTFOBMHPWJTJPO

Slide 46

Slide 46 text

7. ػցֶशͷӡ༻ԽɻΠϯλϏϡʔௐࠪ (ݪจ: Operationalizing Machine Learning: An Interview Study) اۀ͸MLΛӡ༻͢ΔͨΊʹػցֶशΤϯδχΞʢMLEʣʹཔ͍ͬͯ·͢ɻͭ·ΓɺMLύΠϓϥΠϯΛ ຊ൪؀ڥʹಋೖ͠ɺҡ࣋͢Δ͜ͱͰ͢ɻMLΛӡ༻͢ΔϓϩηεʢMLOpsʣ͸ɺʢiʣσʔλͷऩूͱϥ ϕϦϯάɺʢiiʣMLͷύϑΥʔϚϯεΛ޲্ͤ͞ΔͨΊͷ࣮ݧɺʢiiiʣଟஈ֊ͷల։ϓϩηεΛ௨ͯ͠ͷ ධՁɺʢivʣӡ༻தͷύϑΥʔϚϯε௿Լͷ؂ࢹɺͱ͍͏ܧଓతͳϧʔϓͰߏ੒͞Ε͍ͯ·͢ɻMLOps ΛͲͷΑ͏ʹߦ͏ͷ͔ɺະղܾͷ՝୊͸Կ͔ɺͦͯ͠πʔϧϏϧμʔʹͱͬͯͲͷΑ͏ͳҙຯ͕͋Δͷ ͔ʁࢲͨͪ͸ɺνϟοτϘοτɺࣗ཯૸ߦंɺۚ༥ͳͲɺ͞·͟·ͳΞϓϦέʔγϣϯͰ׆༂͢Δ18ਓ ͷMLEʹ൒ߏ଄ԽΤεϊάϥϑΟοΫΠϯλϏϡʔΛ࣮ࢪ͠·ͨ͠ɻΠϯλϏϡʔͰ͸ɺML ͷຊ൪ల ։ͷ੒ޭΛࢧ഑͢Δ 3 ͭͷม਺͕໌Β͔ʹͳΓ·ͨ͠ɻϕϩγςΟɺόϦσʔγϣϯɺόʔδϣχϯά Ͱ͢ɻզʑ͸ɺMLͷ࣮ݧɺσϓϩΠϝϯτɺͦͯ͠ຊ൪ύϑΥʔϚϯεΛҡ࣋͢ΔͨΊͷҰൠతͳϓϥ ΫςΟεΛཁ໿͍ͯ͠·͢ɻ࠷ޙʹɺΠϯλϏϡʔʹ౴͑ͯ͘ΕͨਓͨͪͷϖΠϯϙΠϯτ΍Ξϯνύ λʔϯʹ͍ͭͯɺπʔϧσβΠϯ΁ͷࣔࠦΛؚΊͯٞ࿦͠·͢ɻ w ໨తɿػցֶशͷӡ༻ .-0QT ʹ͍ͭͯͷΠϯλϏϡʔௐࠪ w ੒Ռɿ.-ಋೖͷͨΊͷϓϥΫςΟεΛཁ໿ w ํ๏ɿػցֶशΤϯδχΞ໊΁ͷΠϯλϏϡʔ w ݻ༗໊ɿ w ஶऀॴଐɿΧϦϑΥϧχΞେֶόʔΫϨʔߍ https://arxiv.org/abs/2209.09125v1

Slide 47

Slide 47 text

MLOps ੒ޭͷൿ݃ • ߴ଎ 
 ΞΠσΞͷ࣮ݱ଎౓ɺόάʹਝ଎ʹݕূͰ͖Δσόοά؀ڥΛ༻ҙ ͢ΔͳͲɻ • (ՄೳͳݶΓૣ͍ஈ֊Ͱͷ)ݕূ 
 ॲཧίετΛԼ͛ΔͨΊɺͰ͖Δ͚ͩૣ͘มߋΛςετͯ͠ɺΞΠ σΞΛݕূͯ͠מΓࠐΉɻόά؂ࢹΛ͢Δɻ൓෮αΠΫϧΛߴ଎Խ ͢Δɻ • όʔδϣχϯά(όʔδϣϯຖͷҡ࣋؅ཧ) 
 ੜ࢈ఀࢭ࣌ؒΛ࠷খݶʹ཈͑ΔͨΊɺෳ਺ͷόʔδϣϯͷຊ൪Ϟσ ϧΛҡ࣋ɾ؅ཧɾ੾Γସ͑ΒΕΔ࢓૊ΈΛ༻ҙ͢Δɻ

Slide 48

Slide 48 text

8. ඇ৵ऻత೴ه࿥͔ΒͷԻ੠ղಡ (ݪจ: Decoding speech from non-invasive brain recordings) ೴׆ಈ͔ΒݴޠΛղಡ͢Δ͜ͱ͸ɺϔϧεέΞͱਆܦՊֶͷ྆෼໺ʹ͓͍ͯ଴ͪ๬·Ε͍ͯͨ໨ඪͰ͋Δɻجຊ తͳݴޠ՝୊ʹର͢Δ৵ऻతͳ೴൓Ԡ͔Β܇࿅͞Εͨඃݧऀݻ༗ͷύΠϓϥΠϯ͕ɺղऍՄೳͳಛ௃ʢจࣈɺ୯ ޠɺεϖΫτϩάϥϜͳͲʣΛޮ཰తʹσίʔυ͢ΔΑ͏ʹͳͬͨͷͰ͢ɻ͔͠͠ɺ͜ͷΞϓϩʔνΛࣗવͳԻ ੠΍ඇ৵ऻతͳ೴ه࿥΁֦ு͢Δ͜ͱ͸ɺґવͱͯ͠େ͖ͳ՝୊Ͱ͋Δɻຊ࿦จͰ͸ɺࣗવԻ੠ͷࣗݾڭࢣ෇͖ දݱΛ༧ଌ͢ΔͨΊʹɺେن໛ͳݸਓίϗʔτʹ͓͍ͯରরֶशͰ܇࿅͞Εͨ୯ҰͷΤϯυπʔΤϯυͷΞʔΩ ςΫνϟΛఏҊ͢Δɻզʑ͸ɺ169ਓͷϘϥϯςΟΞ͕ࣗવͳԻ੠Λฉ͖ͳ͕Β೴࣓ਤ΍೴೾Λه࿥ͨ͠4ͭͷ ެ։σʔληοτͰզʑͷϞσϧΛධՁͨ͠ɻͦͷ݁ՌɺզʑͷϞσϧ͸3ඵؒͷMEG৴߸͔Βɺ1,594ݸͷҟ ͳΔηάϝϯτͷ͏ͪ72.5%ͷ্Ґ10Ґ·Ͱͷਫ਼౓ʢ্Ґ1Ґ͸44%ʣͱɺEEGه࿥Ͱ͸2,604ݸͷηάϝϯτ ͷ͏ͪ19.1%ͷਫ਼౓ͰରԠ͢ΔԻ੠ηάϝϯτΛࣝผͰ͖Δ͜ͱ͕ࣔ͞Εͨʢֶ͕ͨͬͯ͠शηοτʹଘࡏ͠ͳ ͍ϑϨʔζΛ෮߸͢Δ͜ͱ͕ՄೳͱͳͬͨʣɻϞσϧൺֱͱΞϒϨʔγϣϯղੳʹΑΓɺ͜ΕΒͷੑೳ͸ɺզʑ ͷઃܭ্ͷબ୒ɺ͢ͳΘͪɺʢiʣରর໨తɺʢiiʣࣄલʹֶशͨ͠Ի੠දݱɺʢiiiʣෳ਺ͷࢀՃऀʹಉ࣌ʹֶश ͤͨ͞ڞ௨ͷ৞ΈࠐΈΞʔΩςΫνϟͷ࢖༻͔Β௚઀తʹརӹΛಘ͍ͯΔ͜ͱ͕ࣔ͞Εͨɻ͜ΕΒͷ݁Ռ͸ɺඇ ৵ऻతͳ೴׆ಈه࿥͔ΒࣗવݴޠॲཧΛϦΞϧλΠϜͰղಡ͢ΔͨΊͷ༗๬ͳಓےΛ໌Β͔ʹ͢Δ΋ͷͰ͋Δɻ w ໨తɿ೴׆ಈ͔Βͷݴޠղಡ w ํ๏ɿԻ੠ͱ೴࣓ਤ .&( ೴ిਤ &&( Λڭࢣσʔλͱͯ͠ɺ೴ͷ׆ಈΛֶश w ੒Ռɿඵͷ.&(&&(͔Βߴ͍ਫ਼౓ͰԻ੠ηάϝϯτΛࣝผͰ͖ͨ w ݻ༗໊ɿ w ஶऀॴଐɿ.FUB https://arxiv.org/abs/2208.12266v1

Slide 49

Slide 49 text

No content

Slide 50

Slide 50 text

No content

Slide 51

Slide 51 text

No content

Slide 52

Slide 52 text

Thank you for coming Ed. ͱௌ͍ͨ࣌ͷඃݧऀ(3ਓ)ͷ೴൓Ԡ

Slide 53

Slide 53 text

9. େن໛ݴޠϞσϧʹΑΔ஧࣮ͳਪ࿦ (ݪจ: Faithful Reasoning Using Large Language Models) ݱ୅ͷେن໛ݴޠϞσϧ(LM)͸ҹ৅తͳ࣭໰Ԡ౴ೳྗΛ͕ࣔ͢ɺͦͷճ౴͸௨ৗɺϞσϧ΁ ͷ1ճͷݺͼग़͠ͷ݁ՌͰ͋Δɻ͜ͷͨΊɺಛʹຊ࣭తʹଟஈ֊ͷ໰୊ʹ͓͍ͯ͸ɺ޷·͘͠ ͳ͍ෆಁ໌ੑΛ൐͍ɺੑೳΛ௿Լͤ͞Δɻ͜ΕΒͷ੍ݶʹରॲ͢ΔͨΊɺզʑ͸ɺҼՌߏ଄ ͕໰୊ͷ࿦ཧߏ଄Λ൓ө͢ΔϓϩηεΛհͯ͠ɺLM͕஧࣮ͳϚϧνεςοϓਪ࿦Λ࣮ߦͰ͖ ΔΑ͏ʹ͢Δํ๏Λࣔ͢ɻຊख๏͸ɺਪ࿦εςοϓΛ࿈࠯తʹ࣮ߦ͢Δɻ֤εςοϓ͸ɺ2ͭ ͷඍௐ੔͞ΕͨLMʢ1ͭ͸બ୒༻ɺ΋͏1ͭ͸ਪ࿦༻ʣͷݺͼग़͔͠Β੒Γɺ༗ޮͳਪ࿦τ ϨʔεΛੜ੒͢Δɻຊख๏͸ɺਪ࿦඼࣭Λ޲্ͤ͞ΔͨΊʹɺਪ࿦τϨʔεͷۭؒΛ௨ͯ͠ ϏʔϜαʔνΛߦ͏ɻզʑ͸ɺଟஈ֊ͷ࿦ཧతਪ࿦ͱՊֶత࣭໰Ԡ౴ʹ͓͚ΔຊϞσϧͷ༗ ޮੑΛ࣮ূ͠ɺ࠷ऴతͳճ౴ਫ਼౓ʹ͓͍ͯϕʔεϥΠϯΑΓ༏Ε͍ͯΔ͜ͱɺ·ͨɺϢʔβ ͕ଥ౰ੑΛνΣοΫͰ͖ΔਓؒతʹղऍՄೳͳਪ࿦τϨʔεΛੜ੒Ͱ͖Δ͜ͱΛࣔ͢ɻ w ໨తɿେن໛ݴޠϞσϧͷ࣭໰Ԡ౴ೳྗͷվྑ w ੒Ռɿଟஈ֊ͷ࿦ཧతਪ࿦ͱՊֶత࣭໰Ԡ౴ʹ͓͚ΔຊϞσϧͷ༗ޮੑΛ࣮ূ w ํ๏ɿਪ࿦εςοϓΛϚϧνεςοϓͰ࣮ߦ͢Δ w ݻ༗໊ɿ w ஶऀॴଐɿ%FFQ.JOE https://arxiv.org/abs/2208.14271v1

Slide 54

Slide 54 text

ContextΛݩʹબ୒ͱਪଌΛ܁Γฦ͠ ͯ࠷ऴతʹQuestionͷ౴͑Λಋ͘ɻ

Slide 55

Slide 55 text

No content

Slide 56

Slide 56 text

No content

Slide 57

Slide 57 text

• SelectionϞσϧ…Context͚͔ͩΒཁૉબ୒ • InferenceϞσϧ…Selection݁Ռ͚͔ͩΒਪ࿦ • HalterϞσϧ…Infarance݁Ռʴ࣭໰͔Β౴͑ Λग़͢ɻ·ͩෆ໌ͳΒUnknownɻ

Slide 58

Slide 58 text

Selection -> Inference -> Halter

Slide 59

Slide 59 text

10. AudioLM: ݴޠϞσϦϯάʹΑΔԻ੠ੜ੒ͷΞϓϩʔν (ݪจ: AudioLM: a Language Modeling Approach to Audio Generation) ຊ࿦จͰ͸ɺ௕ظతͳҰ؏ੑΛอͬͨߴ඼࣭ͳԻ੠ੜ੒ͷͨΊͷϑϨʔϜϫʔΫͰ͋ΔAudioLMΛ঺հ ͢ΔɻAudioLM͸ೖྗԻ੠Λ཭ࢄతͳτʔΫϯͷྻʹରԠ෇͚ɺԻ੠ੜ੒Λ͜ͷදݱۭؒʹ͓͚Δݴޠ ϞσϦϯάλεΫͱͯ͠౤͔͚͛Δɻզʑ͸ɺطଘͷԻ੠τʔΫϯԽث͕࠶ߏ੒඼࣭ͱ௕ظߏ଄ͷؒͰ ͍͔ʹҟͳΔτϨʔυΦϑΛఏڙ͍ͯ͠Δ͔Λࣔ͠ɺ྆໨తΛୡ੒͢ΔͨΊͷϋΠϒϦουτʔΫφΠ θʔγϣϯεΩʔϜΛఏҊ͢Δɻ͢ͳΘͪɺԻ੠Ͱࣄલֶशͨ͠ϚεΫݴޠϞσϧͷ཭ࢄԽ׆ੑΛར༻ ͯ͠௕ظߏ଄Λଊ͑ɺχϡʔϥϧԻ੠ίʔσοΫʹΑͬͯੜ੒͞ΕΔ཭ࢄԽίʔυΛར༻ͯ͠ߴ඼࣭ͳ ߹੒Λ࣮ݱ͢ΔɻAudioLM͸ɺେن໛ͳੜԻ੠ίʔύεͰֶश͢Δ͜ͱͰɺ୹͍ϓϩϯϓτͰࣗવͰҰ ؏ੑͷ͋Δ࿈ଓԻ੠Λੜ੒͢Δ͜ͱΛֶश͢Δɻ·ͨɺॻ͖ى͜͠΍஫ऍͷͳ͍Ի੠ʹରֶͯ͠शΛ ߦͬͨ৔߹ɺAudioLM͸ߏจతɾҙຯతʹଥ౰ͳԻ੠ͷ࿈ଓΛੜ੒͠ɺ͞Βʹະ஌ͷ࿩ऀʹରͯ͠΋࿩ ऀͷಛఆͱӆ཯Λҡ࣋͢Δ͜ͱ͕Ͱ͖Δɻ͞Βʹɺզʑ͸ɺԻָͷ৅௃తͳදݱ͕ͳ͍ʹ΋͔͔ΘΒ ͣɺटඌҰ؏ͨ͠ϐΞϊԻָͷ࿈ଓΛੜ੒͢Δ͜ͱͰɺզʑͷΞϓϩʔν͕Ի੠Λ௒͑Δ͜ͱΛ࣮ূ͢ Δɻ w ໨తɿߴ඼࣭ͳԻ੠ੜ੒Ϟσϧͷ࡞੒ w ੒ՌɿԻ੠ੜ੒ϑϨʔϜϫʔΫ"VEJP-.ͷ։ൃ w ํ๏ɿҙຯτʔΫϯͱԻڹτʔΫϯΛ૊Έ߹Θͤͨ̏ஈ֊ͷݴޠϞσϧΛ઀ଓ w ݻ༗໊ɿ"VEJP-. w ஶऀॴଐɿ(PPHMF3FTFBSDI https://arxiv.org/abs/2209.03143v1

Slide 60

Slide 60 text

ֶशσʔλ͔ΒԻڹτʔΫϯɾ ҙຯτʔΫϯΛ࡞Δ

Slide 61

Slide 61 text

ࢀߟ: SoundStream • ϏοτϨʔτՄมͳχϡʔϥϧΦʔσΟΦ ίʔσοΫ

Slide 62

Slide 62 text

ԻڹτʔΫϯɾҙຯτʔΫϯΛ εςʔδʹΑͬͯ࢖͍෼͚Δ • ҙຯτʔΫϯ: ݴޠత಺༰΍Իָͷટ཯ɾϦζϜ • ߥ͍ԻڹτʔΫϯ: ࿩ऀͷࣝผ΍࿥Ի৚݅ͳͲͷԻڹಛੑ • ਫ਼ີͳԻڹτʔΫϯ: ߴ඼࣭ͷԻ੠߹੒

Slide 63

Slide 63 text

σϞαΠτ IUUQTHPPHMFSFTFBSDIHJUIVCJPTFBOFUBVEJPMNFYBNQMFT

Slide 64

Slide 64 text

σϞͷछྨ • εϐʔνͷ๯಄3ඵͷԻ੠͔ΒࣗવͳܧଓεϐʔνΛੜ੒ • ࿩ऀ͚ͩม͑ͨಉ͡εϐʔνͷੜ੒ • ϥϯμϜͳ࿩ऀɾݴޠɾ؀ڥͰແ৚݅ੜ੒ • ҙຯτʔΫϯͳ͠Ͱͷੜ੒(ҙຯΛͳ͞ͳ͍εϐʔν) • SoundStreamͰͷԻ੠࠶ߏ੒࣌ͷྔࢠԽԻ࣭ൺֱ • ϐΞϊટ཯ͷ๯಄4ඵͷԻ੠͔Βࣗવͳટ཯Λੜ੒

Slide 65

Slide 65 text

ෆਖ਼ར༻Λݕग़͢ΔࢼΈ • ؆୯ʹߴ඼࣭ͳ߹੒Ի੠͕࡞Εͯ͠·͏ͷͰ๷ࢭࡦ Λ༻ҙ(Section IV-H) • ಛఆͷԻ੠σʔλ͕AudioLMͰ࡞ΒΕͨ߹੒σʔλ ͔Ͳ͏͔Λݕग़͢ΔػߏΛ૊ΈࠐΜͩɻ(ݩσʔλͱ ߹੒σʔλͰֶश) • ਓؒͷࣖʹ͸۠ผ͕͔ͭͳͯ͘΋ػցతʹ͸؆୯ʹ ൑ผͰ͖Δ͜ͱ͕Θ͔ͬͨ

Slide 66

Slide 66 text

DeepL Translator (deepl.com) https://www.deepl.com/en/translator