(make girls moe) Yanghua Jin School of Computer Science Fudan University Jiakai Zhang School of Computer Science Carnegie Mellon University Minjun Li School of Computer Science Fudan Univerisity Yingtao Tian Department of Computer Science Stony Brook University Huachun Zhu School of Mathematics Fudan Univerisity Zhihao Fang Department of Architecture Tongji Univerisity ͲΜͳͷʁ ઌߦݚڀͱൺͯͲ͕͍͜͢͝ʁ ٕज़ख๏ͷΩϞͲ͜ʁ ٞ͋Δʁ ࣍ʹಡΉ͖จʁ Ξχϝإը૾ੜɽ"$("/Λ༻͍ͯɼܕϝΨω ͳͲͷಛΛදͤΔɽ ΑΓߴղ૾Խɽ જࡏมΛม͑Δ͜ͱͰࣗࡏʹը૾ੜͰ͖Δɽ (FOFSBUPSʹ%FDPOWPMVUJPOͷΘΓʹ1JYFM 4IVGGMFSΛ͍ɼ433FT/FU MJLFͷϞσϧʹͨ͠ɽ తؔʹϖφϧςΟ߲ΛՃ͑Δ%3"("/Λ༻͍ͨɽ 8FC%//ͰɼXFC্Ͱߴʹը૾ੜ͕Ͱ͖Δɽ Ͳ͏ͬͯ༗ޮͩͱݕূͨ͠ʁ ରԠ͢Δϥϕϧͷը૾͕ੜͰ͖͔ͨࢹͰ֬ೝͨ͠ɽ '*%TDPSFΛൺֱͨ͠ɽ ϥϯμϜϊΠζ෦͕࣭ʹେ͖͘࡞༻͢Δɽϥϕϧ ͕σʔληοτʹۉʹࢄ͞Ε͍ͯͳ͍ͨΊɼϥϕ ϧͷΈ߹ΘͤʹΑͬͯྑͳը૾͕ಘΒΕͳ͍ɽ 130(3&44*7&(308*/(0'("/4'03*.1307&% 26"-*5: 45"#*-*5: "/%7"3*"5*0/ *$-3 େીࠜ ਓؒίʔε
, Toshiki Kataoka1 , Masanori Koyama2 , Yuichi Yoshida 1 Preferred Networks, Inc. 2 Ritsumeikan University 3 National Institute of Informatics ͲΜͳͷʁ ઌߦݚڀͱൺͯͲ͕͍͜͢͝ʁ ٕज़ख๏ͷΩϞͲ͜ʁ ٞ͋Δʁ ࣍ʹಡΉ͖จʁ ("/ͷֶशΛ҆ఆͤ͞ΔͨΊɼॏΈΛTQFDUSBM OPSNBMJ[BUJPOͨ͠ɽ ϋΠύʔύϥϝʔλͱͯ͠ɼϦϓγοπఆΛௐ͢ Δ͚ͩͰྑ͘ͳͬͨɽ·ͨɼௐ͠ͳͯ͘ྑ͍ੑೳ ͕ग़Δɽܭࢉίετখ͍͞ɽ ϨΠϠʔ͝ͱͷॏΈͷεϖΫτϧϊϧϜΛਖ਼نԽ͢Δ͜ͱͰɼ ϨΠϠʔ͝ͱͱɼϨΠϠʔશମͷϦϓγοπ੍ΛͰ͑Β ΕΔɽ Ͳ͏ͬͯ༗ޮͩͱݕূͨ͠ʁ $*'"3ͱ45-ɼ*NBHF/FUͰը૾ੜɽଞͷਖ਼نԽख๏ ͱൺͯੑೳΛൺֱɽ 8FJHIUOPSNBMJ[BUJPOΑΓ΄Ͳ͍ɽ Tim Salimans and Diederik P Kingma. Weight normalization: A simple reparameterization to accelerate train-ing of deep neural networks. In NIPS, pp. 901–909, 2016. େીࠜ ਓؒίʔε
model for speech separation - SIGGRAPH 2018 ARIEL EPHRAT, Google Research and The Hebrew University of Jerusalem, Israel INBAR MOSSERI, Google Research ORAN LANG, Google Research TALI DEKEL, Google Research KEVIN WILSON, Google Research AVINATAN HASSIDIM, Google Research WILLIAM T. FREEMAN, Google Research MICHAEL RUBINSTEIN, Google Research ͲΜͳͷʁ ઌߦݚڀͱൺͯͲ͕͍͜͢͝ʁ ٕज़ख๏ͷΩϞͲ͜ʁ ٞ͋Δʁ ࣍ʹಡΉ͖จʁ ෳͷऀ͔Βɼը૾ͱԻΛͱʹϑΟϧλΛ࡞ ͠ɼ֤ʑͷԻΛநग़Ͱ͖Δɽ "741&&$)%"5"4&5 إಈըͱԻ ͷߏஙɽ Ի͕֤ऀ͝ͱʹಠཱ͍ͯ͠ͳ͍σʔληοτ͔Β ֶश͕Ͱ͖Δɽ ֤ʑͷإͷඵؒͷը૾ʢGSBNF GQT ͱ45'5 TIPSUUFSN ϑʔϦΤม ͨ͠Ի͝ͱʹֶशɽऀ͝ͱͷϚεΫΛग़ྗ ͠ɼ*4*'5ͯ͠Իʹมɽ Ͳ͏ͬͯ༗ޮͩͱݕূͨ͠ʁ TJHOBMUPEJTUPSUJPOSBUJP 4%3 ͱ͍͏ࢦඪΛ༻ɽ ଞͷख๏Ͱऀ͝ͱʹಠཱͨ͠σʔληοτΛ༻͍͍ͯͨʹ ͔͔ΘΒͣɼຊख๏ͷ΄͏্͕ճͬͨɽ إͷಈ͖͕ͲΕ͚ͩد༩͔ͨ͠ੳͨ͠ɽޱҎ֎ॏ ཁΒ͍͠ɽ David F. Harwath, Antonio Torralba, and James R. Glass. 2016. Unsupervised Learning of Spoken Language with Visual Context. In NIPS. େીࠜ ਓؒίʔε
2017 ARIEL EPHRAT, Google Research and The Hebrew University of Jerusalem, Israel INBAR MOSSERI, Google Research ORAN LANG, Google Research TALI DEKEL, Google Research KEVIN WILSON, Google Research AVINATAN HASSIDIM, Google Research WILLIAM T. FREEMAN, Google Research MICHAEL RUBINSTEIN, Google Research ͲΜͳͷʁ ઌߦݚڀͱൺͯͲ͕͍͜͢͝ʁ ٕज़ख๏ͷΩϞͲ͜ʁ ٞ͋Δʁ ࣍ʹಡΉ͖จʁ ޯ߱Լ๏Λ͍ͬͯΔͯ͢ͷख๏ʹ͑ΔNFUB MFBSOJOHख๏.".-ͷఏҊɽ Ϟσϧύϥϝʔλʹ੍ΛՃ͑ͳ͍͍ͯ͘ɽ λεΫू߹ʹ͓͍ͯ'JOF5VOJOH͢Δͱ͍͍ʹߦ͘Α͏ͳॳ ظΛֶशɽ Ͳ͏ͬͯ༗ޮͩͱݕূͨ͠ʁ ճؼσʔλϙΠϯτ͕গͳ͍ͱ͖Ͱɼਖ਼ݭΛֶशՄೳɽ ྨଞͷϝλख๏ͱൺͯߴਫ਼ ڧԽֶशΑΓૣֶ͘शՄೳ ϚϧνλεΫʹର͢ΔॳظԽΛਂֶशڧԽֶशͷ ඪ४తͳཁૉͱ͢Δ͜ͱ͕͜Ε͔ΒॏཁʹͳΓͦ͏ɽ Al-Shedivat, Maruan, et al. "Continuous adaptation via meta- learning in nonstationary and competitive environments." arXiv preprint arXiv:1710.03641 (2017). େીࠜ ਓؒίʔε
of Science and Technology, Yandex Andrea Vedaldi University of Oxford [email protected] Victor Lempitsky Skolkovo Institute of Science and Technology (Skoltech) [email protected] ͲΜͳͷʁ ઌߦݚڀͱൺͯͲ͕͍͜͢͝ʁ ٕज़ख๏ͷΩϞͲ͜ʁ ٞ͋Δʁ ࣍ʹಡΉ͖จʁ ʮը૾ͱ͜͏͍͏ͷʯͱ͍͏ࣄલใ͕$//ͷߏࣗ ମʹͦͦඋΘ͍ͬͯΔͱԾఆ͠ɼ୯Ұը૾ͷΈΛ༻͍ͨ ֶशͰɼͦͷը૾ͷEFOPJTJOH IJHISFTPMVUJPO JOQBJOUJOH Λ࣮ݱɽ େྔͷσʔληοτ͕͍Βͳ͍ɽ ϥϯμϜϊΠζ͔Βɼ៉ྷʹ͍ͨ͠ը૾Λۙࣅ͢ΔΑ͏ʹੜ ͤ͞Δͱɼӄʹ ࣜͷ੍͕ຬͨ͞Εɼ͖Ε͍ͳը૾͕ੜ ͞ΕΔɽʢϒϥοΫϘοΫεײʣ Ͳ͏ͬͯ༗ޮͩͱݕূͨ͠ʁ EFOPJTJOHͰ14/3͕405"ͱಉఔ ߴղ૾ԽͰ14/3͕405"ͱಉఔ 4FU<>ͱ4FU <>σʔληοτͰͱ JOQBJOUͰ $//ͷߏ্ɼ("/ͱಉ༷ɼ֨ࢠঢ়ʹͳΓ͍͢ʁ [24] V. Papyan, Y. Romano, and M. Elad. Convolutional neural networks analyzed via convolutional sparse coding. Journal of Machine Learning Research, 18(83):1–52, 2017. େીࠜ ਓؒίʔε
Berkeley Trevor Darrell UC Berkeley Richard Zhang UC Berkeley Alexei A. Efros Oliver Wang Deepak Pathak UC Berkeley Eli Shechtman Adobe Research UC Berkeley Adobe Research ͲΜͳͷʁ ઌߦݚڀͱൺͯͲ͕͍͜͢͝ʁ ٕज़ख๏ͷΩϞͲ͜ʁ ٞ͋Δʁ ࣍ʹಡΉ͖จʁ ը૾ͷը෩มΛύλʔϯͰͳ͘ෳͰ͖Δɽ QJYQJYͰੜ࣌ʹ༩͑ΔTFFEϊΠζΛͯ͠ NPEFDPMMBQTFͯ͠͠·͕ͬͨɼ͠ͳ͘ͳͬͨɽ D7"&("/ͱD-3("/Λ·ͱΊͨɼ#JDZDMF("/ͱ͍͏Ϟσ ϧɽ Ͳ͏ͬͯ༗ޮͩͱݕূͨ͠ʁ edges → photos , Google maps → satellite, labels → images, and outdoor night → day images ͷมΛpix2pix+noise, cAE-GAN, cVAE-GAN, cVAE-GAN++, cLR-GAN, BicycleGAN ͰͬͯΈͨɽGoogle maps → satellites ͰLPIPSͱ͍͏ଟ༷ੑΛଌΔείΞΛൺֱɽ Amazon ϝΧχΧϧλʔΫͰὃ͞ΕͨΛൺֱɽ ଟ༷ੑΛଌΔ-1*14ͱ͍͏ࢦඪͷείΞը૾͕ෆࣗવͳͱ͖ ߴ͘ͳΔɽ MBUFOUͷ࣍ݩগͳ͗͢ΔͱNPEFDPMMBQTF͢Δ͕ɼଟ͗͢Δͱ ΑΓଟ͘ͷใྔΛ࣋ͬͯ͠·͏ͨΊྑ͘ͳ͍ɽ • A. B. L. Larsen, S. K. Sønderby, H. Larochelle, and O. Winther. Autoencoding beyond pixels using a learned similarity metric. In ICML, 2016. େીࠜ ਓؒίʔε
2016 Anders Boesen Lindbo Larsen1 Søren Kaae Sønderby2 Hugo Larochelle3 Ole Winther1,2 1 Department for Applied Mathematics and Computer Science, Technical University of Denmark 2 Bioinformatics Centre, Department of Biology, University of Copenhagen, Denmark 3 Twitter, Cambridge, MA, USA ͲΜͳͷʁ 7"&ͱ("/ΛΈ߹Θͤͨɽ("/ͷEJTDSJNJOBUPSͰ ֶश͞ΕͨಛදݱΛ7"&ͷPCKFDUJWFʹ༻Ͱ͖Δɽ େીࠜ ਓؒίʔε Vincent Dumoulin1, Ishmael Belghazi1, Ben Poole2 Olivier Mastropietro1, Alex Lamb1, Martin Arjovsky3 Aaron Courville1† 1 MILA, Université de Montréal, fi[email protected]. 2 Neural Dynamics and Computation Lab, Stanford, [email protected]. 3 New York University, [email protected]. †CIFAR Fellow. ͲΜͳͷʁ Y͔Β(FOFSBUPSʹΑͬͯਪͨ͠[ͱɼ[͔Β (FOFSBUPSʹΑͬͯਪͨ͠ϖΞಉ࢜Λ%JTDSJNJOBUPS ʹΑͬͯൺΔɽ ADVERSARIALLY LEARNED INFERENCE - ICLR 2017
- NIPS 2016 • X. Chen, Y. Duan, R. Houthooft, J. Schulman, I. Sutskever, and P. Abbeel. ͲΜͳͷʁ ڭࢣͳֶ͠शͰੜը૾Λ੍ޚͰ͖Δ જࡏมͱ؍ͷؒͷ૬ޓใྔΛ࠷େԽ͢Δ େીࠜ ਓؒίʔε • R. Zhang, J.-Y. Zhu, P. Isola, X. Geng, A. S. Lin, T. Yu, and A. A. Efros. ͲΜͳͷʁ നࠇࣸਅΛϦΞϧλΠϜͰண৭Ͱ͖ΔɽϢʔβʔΠϯ ϓοτܕͱσʔλυϦϒϯܕͷண৭ख๏ͷ༥߹ɽ ૄͳೖྗ͔ΒશମΛਪଌɽ దͨ͠ΧϥʔύϨοτΛαδΣετɽ Ͱྑ͍ண৭͕Ͱ͖Δ͔࣮ݧɽ VOVTVBMͳண৭ੜͰ͖Δɽ Real-Time User-Guided Image Colorization with Learned Deep Priors - SIGGRAPH 2017
Yang Chen Tsinghua University, China Yu-Kun Lai Cardiff University, UK [email protected] Yong-Jin Liu∗ Tsinghua University, China ͲΜͳͷʁ ઌߦݚڀͱൺͯͲ͕͍͜͢͝ʁ ٕज़ख๏ͷΩϞͲ͜ʁ ٞ͋Δʁ ࣍ʹಡΉ͖จʁ ࣸਅΛອը෩ʹม͢Δ$BSUPPO("/ΛఏҊɽ ରσʔλ͕ඞཁͳ͍ɽ ଛࣦؔʹɼ௨ৗͷࣝผଛࣦʹՃ͑ͯɼֶशࡁΈ7((͔Βͷग़ ྗΛ༻͍ͯ(͕ੜ͢ΔֆͷҙຯΛࣦΘͳ͍Α͏͢Δϩε ͱɼ ΤοδΛࡍཱͨͤΔϩε ී௨ͷࣸਅ͔ɼΤοδΛऔΓআ͍ͨ ອը͔ɼΤοδͦͷ··ͷອը͔ɼΛࣝผ Λ࡞ͬͨ Ͳ͏ͬͯ༗ޮͩͱݕূͨ͠ʁ ৽ւ෩ɼٶ࡚ॣ෩ͳͲʹมɽNSTɼCycleGANͱൺֱɽ ΑΓອըͬΆ͘ɼΤοδࡍཱͭΑ͏ʹͳͬͨɽ কདྷతʹਓըɼಈըʹద༻͍ͨ͠ɽ • [38] J.-Y.Zhu,T.Park,P.Isola,andA.A.Efros. Unpairedimage- to-image translation using cycle-consistent adversarial net- works. In International Conference on Computer Vision, 2017. • [6] L. Gatys, A. Ecker, and M. Bethge. Image style transfer using convolutional neural networks. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2414–2423, 2016. େીࠜ ਓؒίʔε
2017 J.-Y.Zhu,T.Park,P.Isola,andA.A.Efros. ͲΜͳͷʁ $ZDMF("/ ରͷը૾ϖΞ͕ͳͯ͘ը૾ͷม͕Ͱ͖Δɽ େીࠜ ਓؒίʔε L. Gatys, A. Ecker, and M. Bethge. ͲΜͳͷʁ 7((ͰऔΓग़ͨ͠ϨΠϠʔ͝ͱͷTUZMFදݱͱDPOUFOU දݱΛݩʹɼͦΕͧΕΛอ࣋͢ΔΑ͏ͳϩεؔΛ࡞ Γɼը෩ม͢Δɽ Image style transfer using convolutional neural networks - CVPR 2016
2013 A. L. Maas, A. Y. Hannun, and A. Y. Ng. ͲΜͳͷʁ -FBLZ3F-6ͷఏҊɽ ԻೝࣝγεςϜͰTJHNPJEܥͳͲͱൺֱͯ͠ˋྑ͘ ͳͬͨɽ େીࠜ ਓؒίʔε • Y. Chen, Y.-K. Lai, and Y.-J. Liu. ͲΜͳͷʁ άϨΠεέʔϧͰɼࣸਅΛອը෩ʹม͢Δɽ (BUZͷٕज़Λ୯७ʹອըʹ༻ͨ͠ͷʹՃ͑ͯɼࣸਅ ͔ອըՈΛࣝผ͢ΔΑ͏ʹϩεΛՃ͑ͨɽ Transforming photos to comics using convolutional neural networks. - ICIP 2017
2017 J. Liao, Y. Yao, L. Yuan, G. Hua, and S. B. Kang. ͲΜͳͷʁ lEFFQJNBHFBOBMPHZzͱ͍͏ɼҙຯతʹ͍ۙΑ͏ʹ ม͢ΔΑ͏ͳΈΛఏҊɽ SFDPOTUSVDUJPOʹ1BUDI.BUDIΛ࠾༻ͨ͠ɽ େીࠜ ਓؒίʔε
van den Oord et al. ͲΜͳͷʁ લ·ͰͷϐΫηϧ͔Β࣍ͷϐΫηϧΛ༧ଌ͢Δ 1JYFM$//ʹʮ CMJOETQPU ʯΛઃ͚ͨ(BUFE 1JYFM$//ͱɼը૾ͷ݅Ϟσϧ͕Ͱ͖Δ$POEJUJPOBM 1JYFM$//ΛఏҊɽ7"&ͷEFDPEFͷ࣭Λ্ɽ େીࠜ ਓؒίʔε A. R. Zamir et al. ͲΜͳͷʁ %0'Χϝϥͷϙʔζਪఆ ࣄલͷಛྔͳ͠ʹɼGFBUVSFNBUDIJOHͷ405" Generic 3D representation via pose estimation and matching - ECCV 2016
D. J. Rezende et al ͲΜͳͷʁ ը૾͔Β%ͷମΛੜɽ %ີϞσϦϯάΛॳΊͯߦͬͨɽ େીࠜ ਓؒίʔε A. A. Rusu et al ͲΜͳͷʁ Ϩϕϧͷࢹ֮ಛΛߴϨϕϧͷϙϦγʔʹҠ͍ͯ͠ ͘QSPHSFTTJWFOFUXPSLʹΑͬͯγϛϡϨʔγϣϯͰ ͷֶशΛ࣮ڥʹҠ͢ɽ Sim-to-real robot learning from pixels with progressive nets. - arXiv 2016