Upgrade to Pro — share decks privately, control downloads, hide ads and more …

TGS Salt Identification Challenge 12th place solution

5d07c24d8a7903a287556c5e76c0609b?s=47 catla
November 30, 2018
Research
3
8.8k

TGS Salt Identification Challenge 12th place solution

5d07c24d8a7903a287556c5e76c0609b?s=128

catla

November 30, 2018
Tweet

More Decks by catla

See All by catla
ベイズ深層学習(6.3)
5d07c24d8a7903a287556c5e76c0609b?s=48 catla
1
140
ベイズ深層学習(6.2)
5d07c24d8a7903a287556c5e76c0609b?s=48 catla
3
140
[読み会資料] Federated Learning for Vision-and-Language Grounding Problems
5d07c24d8a7903a287556c5e76c0609b?s=48 catla
0
86
ベイズ深層学習(5.1~5.2)
5d07c24d8a7903a287556c5e76c0609b?s=48 catla
0
130
ベイズ深層学習(4.1)
5d07c24d8a7903a287556c5e76c0609b?s=48 catla
0
250
ベイズ深層学習(3.3~3.4)
5d07c24d8a7903a287556c5e76c0609b?s=48 catla
17
9.7k
ベイズ深層学習(2.2~2.4)
5d07c24d8a7903a287556c5e76c0609b?s=48 catla
6
1k
23回アルゴリズムコンテスト 1位解法
5d07c24d8a7903a287556c5e76c0609b?s=48 catla
6
580
Learning Lightweight Lane Detection CNNs by Self Attention Distillation(ICCV2019)の紹介
5d07c24d8a7903a287556c5e76c0609b?s=48 catla
0
360

Other Decks in Research

See All in Research
ABCIで回そう爆速深層学習 (基礎編)
E30ae1fd7ecce7aed0032ea0f87f6520?s=48 yoshipon
8
4.4k
時間情報表現抽出とルールベース解析器のこれから / Temporal Expression Analysis in Japanese and Future of Rule-based Approach
64fb973c47087ece7fb9b45d5b8593a4?s=48 yag_ays
PRO
0
730
ECサイトにおけるデータ駆動型意思決定のための非線形テンソル因子分解を用いたVisual Analyticsシステムの検討
B0edf7a7acbd634903de0106b1743056?s=48 ae14watanabe
0
200
ICCV 2021 ただただVision and Languageの論文を読んでみた
Be0f86176276318b4b9775d795278f7e?s=48 yushiku
3
1.7k
第8回チャンピオンズミーティング・サジタリウス杯決勝集計 / Umamusume Sagittarius 2021 Final
9782dd351a8f1737d4ef02d839e821f9?s=48 kitachan_black
0
1.1k
第10回チャンピオンズミーティング・アクエリアス杯ラウンド2集計 / Umamusume Aquarius 2022 Round2
9782dd351a8f1737d4ef02d839e821f9?s=48 kitachan_black
0
1.1k
論文読み会資料 When Shift Operation Meets Vision Transformer: An Extremely Simple Alternative to Attention Mechanism
E054bbb78a219fbddbac0bc492aed606?s=48 sensetime_japan
1
720
Recent Findings on Density-Ratio Approaches in Machine Learning
Bb6c3fc8c577710c72d03aeb4fa56bf6?s=48 masakat0
0
220
測定に起因するバイアスについて
712e4ea459afe9662a85fd972798026c?s=48 yurika_kawazoe
1
1.5k
幼少期の自然体験が理科学習への態度に及ぼす影響
4e437713540d4a32369b9c9f1b624e9c?s=48 arumakan
0
820
Natural language processing tells us the shape of language
0fee20d8bbb7283e1887e7075f638f59?s=48 eumesy
0
260
文献紹介：Dynamic Contextualized Word Embeddings
9e650916f36300d64c9c61eeb4ab697e?s=48 a1da4
2
150

Featured

See All Featured
The Straight Up "How To Draw Better" Workshop
Aff5641764408271f7bc398f2097edd0?s=48 denniskardys
225
120k
How to Ace a Technical Interview
2f5463832ccb768ccb4a1ca3607c27ef?s=48 jacobian
265
21k
Infographics Made Easy
282d599b414022eba5096510f675b6e2?s=48 chrislema
233
17k
Thoughts on Productivity
A16eb159db895e3b01d3dc95767ad595?s=48 jonyablonski
43
2.2k
Teambox: Starting and Learning
Aebc2d07a50011e2a30d38435fde564a?s=48 jrom
121
7.6k
StorybookのUI Testing Handbookを読んだ
50fc0fdc2327ecbe7350645812de52e3?s=48 zakiyama
4
2k
The Most Common Mistakes in Cover Letters
1ce0eb06fd6a9e9566a0cb310cfee635?s=48 jrick
PRO
4
24k
Visualization
170b760e0147792d0140e59ec78e9893?s=48 eitanlees
124
11k
Dealing with People You Can't Stand - Big Design 2015
44b54d2ffcb7857004071cc6795dde11?s=48 cassininazir
350
21k
Making the Leap to Tech Lead
32de0bd2ba869609d26fd052a4622778?s=48 cromwellryan
113
6.9k
Clear Off the Table
Ef32e0b1ac5082450dc8c1fca3a26b5c?s=48 cherdarchuk
79
280k
Creatively Recalculating Your Daily Design Routine
48c098b6579220a67a6ba949be6e4b5a?s=48 revolveconf
205
10k

Transcript

  1. 5(44BMU*EFOUJpDBUJPO$IBMMFOHF 5FBN-FBSOJOHUIF'VUVSF UIQMBDFTPMVUJPO DBUMB ,BHHMFNFFUVQ LBHHMF@UPLZP

  2. ࣗݾ঺հ ໊લܡɹঘً   5XJUUFS!LBUTVSB@KQ  ॴଐஜ೾େֶɹ৘ใֶ܈ɹ৘ใՊֶྨɹ೥ ϓϩάϥϛϯάྺ໿೥൒ ઐ߈෼໺ແ͠ ͦͷଞ


    ࠓճ͕,BHHMFॳ௅ઓ

  3. νʔϜ঺հ

  4. ίϯϖͷ֓ཁ ஍਒ը૾͔ΒԘؚ͕·Ε͍ͯΔ෦෼ΛృΓֆ͢Δίϯϖ ηϚϯςΟοΫηάϝϯςʔγϣϯ *OQVUσʔλάϨʔεέʔϧը૾ Y σʔλ EFQUIσʔλը૾ͷਂ͞৘ใ 5SBJOJOHEBUBTFUຕ 5FTUEBUBTFUຕ ࢀՃνʔϜ਺

     UFBNT σʔλαΠζ͕খ͍͞ ظؒɿd

  5. wධՁࢦඪ w *P6 JOUFSTFDUJPOPWFSVOJPO ίϯϖͷ֓ཁ w೉͔ͬͨ͠఺ wॎʹਨ௚ͳϚεΫͷ༧ଌਫ਼౓

  6. 1VCMJD-#UI

  7. 1SJWBUF-#UI

  8. 4PMVUJPO 6/FUϕʔε

  9. 4PMVUJPO

  10. 4PMVUJPO 6/FU

  11. 4PMVUJPO d#BTF&ODPEFSd DBUMB  3FT/F9U MZBLB  3FT/FU BMVQ 

    3FT/FU  3FT/F9U DPTUBT  3FT/FU PVSBOPT  %FOTF/FU USJBO  3FT/FU

  12. 4PMVUJPO .BY1PPMJOH $POWE TUSJEFT *OQVUTJ[F Y Y

  13. 4PMVUJPOd&ODPEFSd #BDLCPOF#MPDL #BDLCPOF#MPDL Y   D   D

    I X D (MPCBMQPPMJOH %FOTF 3FMV %FOTF 4JHNPJE

  14. 4PMVUJPOd&ODPEFSd #BDLCPOF#MPDL #BDLCPOF#MPDL Y Y I X D (MPCBMQPPMJOH %FOTF

      D   D 3FMV %FOTF 4JHNPJE $POWE LFSOFMTJ[F    PVUDIBOOFM 4JHNPJE I X  I X D

  15. 4PMVUJPOdDFOUFSd $POWE #/ 3FMV $POW % $POWE #/ 3FMV $POWE

    #/ 3FMV $POW % $POW % $POW % %%FMBUJPO

  16. 4PMVUJPOd%FDPSEFSd $POWE #/ 3FMV $POWE #/ 3FMV Y Y 

     D   D $POWE LFSOFMTJ[F    PVUDIBOOFM 4JHNPJE I X  I X D $POWE  #/ 3FMV $POWE  #/ 3FMV I X D (MPCBMQPPMJOH %FOTF 3FMV %FOTF 4JHNPJE

  17. 4PMVUJPO -PTT 'PDBM-PTT -PWBT[MPTT BDUJWBUJPO3FMVPSFMV  0QUJNJ[FS "EBN NPNFOUVN4(%

  18. 4PMVUJPO )ZQFSDPMNO

  19. 4PMVUJPO ɾQTFVEPMBCFMJOH ɾ4OBQ4IPU&OTFNCMJOH ɾ'BTU(FPNFUSJD&OTFNCMJOH

  20. 4PMVUJPO

  21. 4PMVUJPO -PTT

  22. 4PMVUJPOd1PTU1SPDFTTJOHd

  23. 4PMVUJPOd1PTU1SPDFTTJOHd

  24. 4PMVUJPOd1PTU1SPDFTTJOHd 1VCMJD-#  1SJWBUF-#ʹ͸ɺ ӨڹΛ༩͑ͳ͔ͬͨ

  25. ͓ΘΓʹ 5XJUUFS աڈͷίϯϖ 2JJUB (JU)VC .FEJVN 4MJEFTIBSF BS9JW ຊ FUDʜ

    ͲΜͲΜ׆༻͠Α͏ʂ ᩦཉʹ୳ͤ͹༗༻ͳ৘ใ͕͝Ζ͝Ζམͪͯ ·͢ɻࠓճͷίϯϖͰ͸ɺ͜ΕΒΛಛʹ׆ ༻͍ͯ͠·ͨ͠ɻͥͻࢀߟʹ