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

꿈꾸는 Agent

Wonseok Jung
September 28, 2018
Research
2
97

꿈꾸는 Agent

Wonseok Jung

September 28, 2018
Tweet

More Decks by Wonseok Jung

See All by Wonseok Jung
Ai for business -self car driving
wonseokjung
0
170
reinforcement_learning_.pdf
wonseokjung
2
1.5k
원석이의 모두연에서 강화학습 보석되기
wonseokjung
0
390
NeuralIPS
wonseokjung
0
300
Introduction Deep Reinforcement Learning
wonseokjung
0
120
Deep reinforcemenet learning -2
wonseokjung
0
160
Deep Reinforcement Learning - Introduction
wonseokjung
1
600
How to become a datascientist ?
wonseokjung
2
2.3k
Review of Taylor series
wonseokjung
1
110

Other Decks in Research

See All in Research
Introduction of NII S. Koyama's Lab (AY2024)
skoyamalab
0
120
20240127_熊本から今いちど真面目に都市交通～めざせ「車1割削減、渋滞半減、公共交通2倍」～ 全国路面電車サミット2024宇都宮
trafficbrain
1
670
ランサーズエージェント_フリーランスエンジニアの年収・キャリアの実態調査2024
lancers_pr
0
110
自己教師あり学習による事前学習（CVIMチュートリアル）
naok615
2
1.4k
オープンな日本語埋め込みモデルの選択肢 / Exploring Publicly Available Japanese Embedding Models
nttcom
14
5.6k
First Authorに俺はなるっ！! IROS’23 CCC2023 FY
shota_nishiyama
0
180
東工大Swallowプロジェクトにおける大規模日本語Webコーパスの構築
aya_se
13
6.7k
媒介分析と疫学
kingqwert
0
100
クロスモーダル表現学習の研究動向: 音声関連を中心として
ryomasumura
3
610
センサデータを活用した 肌質改善への支援システムに関する研究
comfortdesignlab
0
170
3D Human Mesh Estimationについていくつかまとめてみた / Survey about 3D Human Mesh Estimation
nttcom
0
220
第14回対話システムシンポジウム EMNLP 2023 参加報告
atsumoto
0
160

Featured

See All Featured
How GitHub (no longer) Works
holman
305
140k
No one is an island. Learnings from fostering a developers community.
thoeni
16
2.1k
Agile that works and the tools we love
rasmusluckow
325
20k
Fashionably flexible responsive web design (full day workshop)
malarkey
398
65k
How to name files
jennybc
65
93k
Making the Leap to Tech Lead
cromwellryan
125
8.5k
It's Worth the Effort
3n
180
27k
A designer walks into a library…
pauljervisheath
201
23k
Keith and Marios Guide to Fast Websites
keithpitt
408
22k
Intergalactic Javascript Robots from Outer Space
tanoku
266
26k
[RailsConf 2023] Rails as a piece of cake
palkan
27
4k
CoffeeScript is Beautiful & I Never Want to Write Plain JavaScript Again
sstephenson
155
14k

Transcript

  1. ԷԲח"HFOU 8POTFPL+VOH

  2. 
 8POTFPL+VOH $JUZ6OJWFSTJUZPG/FX:PSL#BSVDI$PMMFHF %BUB4DJFODF.BKPS  $POOFYJPO"*'PVOEFS %FFQ-FBSOJOH$PMMFHF3FJOGPSDFNFOU-FBSOJOH3FTFBSDIFS .PEVMBCT$53--FBEFS 4DIPPMPG"*0QFSBUPS
 3FJOGPSDFNFOU-FBSOJOH

    0CKFDU%FUFDUJPO $IBUCPU (JUIVC IUUQTHJUIVCDPNXPOTFPLKVOH 'BDFCPPL IUUQTXXXGBDFCPPLDPNXTKVOH
 #MPH IUUQTXPOTFPLKVOHHJUIVCJP
 :PVUVCF
 IUUQTXXXZPVUVCFDPNDIBOOFM6$N5Y8,EIM8W+6GS3X


  3. "XBSET  Ӗ۽ߥ҃ઁޙചߊ੹؀࢚ "*ӝࣿഄनࠗޙ؀࢚࢚ࣻ Ҵഥ੄ਗ੢಴ହ੢ࣻৈ 
 .PEVMBCT'JSTUDMBTTJGJDBUJPODPNQFUJUJPOT SE
 .PEVMBCT'JSTU3-$PNQFUJUJPOT
 TU


    
 .PEVMBCT4FDPOE3-$PNQFUJUJPOT TU 0QFO"*3FJOGPSDFNFOUMFBSOJOH3FUSP 5PQ

  4. 
 1SFTFOUBUJPOBOE-FDUVSFT  5FOTPSGMPX,33FJOGPSDFNFOU-FBSOJOH ऱ੉Ӓېݠ%*:ъച೟ण ա݅੄गಌܻ݃য়ڣٯڣٯٜ݅ӝ  IUUQTXXXZPVUVCFDPNXBUDI W#.*JSCW2X 

    ী੉੹౟ܳࠗఌ೧5IFSJTFPG3FJOGPSDFNFOU-FBSOJOH IUUQTXXXZPVUVCFDPNXBUDI WVI9V"L3G.2 TUೣԋೞח٩۞׬ஶಌ۠झ3VOOJOH"* IUUQTXXXTMJEFTIBSFOFUXPOTFPLKVOHEFFQMFBSOJOHDPOG ౵੉௑௏ܻইੋҕ૑מगಌܻ݃য়੄Ѣ੄ݽٚѪ IUUQTTQFBLFSEFDLDPNXPOTFPLKVOHBEPUJTVQFSNBSJP XJUISFJOGPSDFNFOUMFBSOJOH  ೠҴ੿ࠁ୊ܻ೟ഥъച೟ण੉ۿҗपઁ IUUQTXXXTMJEFTIBSFOFUXPOTFPLKVOHSM ా҅աۄъച೟ण੉ۿҗपઁ IUUQTXXXTMJEFTIBSFOFUXPOTFPLKVOHSM
  5. None
  6. None

  7. https://medium.com/@madanram/you-only-look-once-aaf841df2c7b

  8. http://www.relationclientmag.fr/Thematique/techno-ux-1256/Breves/Chatbot-B2B-qualite-service-peut-aussi-passer-machines-318813.htm#myUmcA7SSRYBzhzk.97

  9. None

  10. WHAT I’VE DONE REINFORCEMENT LEARNING

  11. R.L REINFORCEMENT LEARNING Reinforcement Learning

  12. (PBM 8BMM "HFOU "DUJPO "DUJPO "DUJPO REINFORCEMENT LEARNING CURRICULUM LEARNING

  13. IMITATION LEARNING *NJUBUJPO-FBSOJOH 5FBDIFS 4UVEFOU REINFORCEMENT LEARNING

  14. R.L REINFORCEMENT LEARNING https://ubisafe.org/explore/demeanure-clipart-prior-knowledge/ Meta Reinforcement Learning Exploration Reinforcement Learning

  15. R.L REINFORCEMENT LEARNING https://ubisafe.org/explore/demeanure-clipart-prior-knowledge/ Meta Reinforcement Learning Exploration Reinforcement Learning

  16. R.L REINFORCEMENT LEARNING https://ubisafe.org/explore/demeanure-clipart-prior-knowledge/ Meta Reinforcement Learning Exploration Reinforcement Learning

  17. HOW ANIMALS LEARN

  18. ALL ANIMALS HAVE THE ABILITY TO LEARN - ݽٚزޛ਷೟णמ۱੉੓׮ -

    ৈѐ੄न҃ࣁನ݅ਸыҊ੓ח৘ࢂԘ݃ࢶ୽ژೠ೟णמ۱੉੓׮ - ݠ୍ܻࣻ߈ࢎIFBEXJUIESBXTSFGMFYਤ೷ೠޛ୓о੓ਸѪ੉ۄ౸ױীٮܲ߈ࢎ೯ز - ৘ࢂԘ݃ࢶ୽੄ݠܻܳѤܻ٘ݶੌ੿Ѣٍܻܳ۽р׮ HOW ANIMALS LEARN

  19. HABITUATION HOW ANIMALS LEARN 'JSTUUSZ 4FDPOEUSZ 5IJSEUSZ

  20. LAW OF EFFECT - &EXBSE5IPSOEJLF   - -BXPGFGGFDUযڃ೯ز੄Ѿҗо݅઒झ۞਋ݶ׮਺ীبӒ೯زਸ߈ࠂೠ׮ ߈؀۽݅઒ೞ૑ঋਵݶӒ೯زਸೞ૑ঋח׮

    - 3FJOGPSDFNFOU ъച ੉੹ীੌযդ೯زਸ߈ࠂೞѱ݅٘ח੗ӓ - 1VOJTINFOU ୊ߥ ੉੹ীੌযդ೯زਸೖೞѱ݅٘ח੗ӓ HOW ANIMALS LEARN

  21. EXAMPLE OF THE LAW OF EFFECT HOW ANIMALS LEARN

  22. HOW HUMANS LEARN

  23. INTERACTION WITH ENVIRONMENT &OWJSPONFOU &YQFSJFODF -FBSO *OUFSBDUJPO HOW HUMANS LEARN

  24. HOW HUMANS LEARN? - 3FJOGPSDFNFOU੉੹ীੌযդ೯زਸ߈ࠂೞѱ݅٘ח੗ӓ - 1VOJTINFOU੉੹ীੌযդ೯زਸೖೞѱ݅٘ח੗ӓ HOW HUMANS LEARN

  25. REINFORCEMENT LEARNING

  26. REINFORCEMENT LEARNING &OWJSPONFOU &YQFSJFODF -FBSO REINFORCEMENT LEARNING *OUFSBDUJPO

  27. LEARNING 3FJOGPSDFNFOUMFBSOJOH਷3FXBSE ࠁ࢚ ਸ୭؀ചೞחBDUJPO ೯ز ਸࢶఖೠ׮ -FBSOFS ߓ਋ח੗ חৈ۞BDUJPOਸ೧ࠁݴ SFXBSEܳо੢֫ѱ߉חBDUJPOਸ଺ח׮

    ࢶఖػBDUJPO੉׼੢੄SFXBSEࡺ݅ইצ ׮਺੄࢚ടژח׮਺ੌযաѱؼ SFXBSEীب৔ೱਸՙசࣻب੓׮ "DUJPO ׼੢੄ ࢚ട߸ച ޷ې੄࢚ട 3FXBSE ޷ې੄3FXBSE REINFORCEMENT LEARNING

  28. MARKOV DECISION PROCESS "DUJPO "HFOU &OWJSPONFOU 3FXBSE A t R

    t 4UBUF S t R t+1 S t+1 REINFORCEMENT LEARNING

  29. AGENT "DUJPO "HFOU &OWJSPONFOU 3FXBSE A t R t 4UBUF

    S t R t+1 S t+1 REINFORCEMENT LEARNING

  30. ACTION "DUJPO "HFOU &OWJSPONFOU 3FXBSE A t R t 4UBUF

    S t R t+1 S t+1 REINFORCEMENT LEARNING

  31. OBSERVATION, REWARD "DUJPO "HFOU &OWJSPONFOU 3FXBSE A t R t

    4UBUF S t R t+1 S t+1 REINFORCEMENT LEARNING

  32. MARKOV DECISION PROCESS &OWJSPONFOU 3FXBSE A t R t S

    t R t+1 S t+1 REINFORCEMENT LEARNING &TDBQFGSPN UIFCPY 1PTJUJWF3FXBSE


  33. MARKOV DECISION PROCESS "DUJPO "HFOU &OWJSPONFOU 3FXBSE A t R

    t 4UBUF S t R t+1 S t+1 SUPERMARIO WITH R.L 3FXBSE  1FOBMUZ

  34. RETURN 4UBUFWBMVF REINFORCEMENT LEARNING 4UBUF"DUJPOWBMVF

  35. EXPLORATION

  36. https://www.youtube.com/watch?v=8vNxjwt2AqY

  37. EXPLOITATION AND EXPLORATION REINFORCEMENT LEARNING Exploitation Exploration ?

  38. EXPLOITATION REINFORCEMENT LEARNING Exploitation Exploration ?

  39. EXPLORATION REINFORCEMENT LEARNING Exploitation ? Exploration

  40. IMPORTANCE OF EXPLORATION ࣄ੉ 3VTTJBO#MVF ࢓ $VSJPTJUZ
 ಽ੉ .VODILJO ࢓

    'PPE REINFORCEMENT LEARNING

  41. IMPORTANCE OF EXPLORATION-2 REINFORCEMENT LEARNING ಽ੉ ࣄ੉ ;FSP
 FYQMPSBUJPO &YQMPSBUJPO

  42. IMPORTANCE OF EXPLORATION-3 REINFORCEMENT LEARNING ಽ੉ ࣄ੉ 'BJM

  43. GOAL SUPERMARIO WITH R.L

  44. None

  45. ALL WORLDS AND LEVELS SUPERMARIO WITH R.L env = gym_super_mario_bros.make('SuperMarioBros-<world>-<level>-v<version>')

                   

  46. APPROXIMATE ACTION-VALUE SUPERMARIO WITH R.L

  47. DOUBLE DQN SUPERMARIO WITH R.L JOQVU "DUJPO WBMVF &OW 2/FUXPSL

    s’ s 3FQMBZNFNPSZ 2 T B a r (S t , A t , R t+ 1 , S t+ 1 )

  48. 1000EPISODES, 3000EPISODES, TRAINING SUPERMARIO WITH R.L FQJTPEFT FQJTPEFT

  49. 5000 EPISODE SUPERMARIO WITH R.L FQJTPEFT %BZT

  50. ALPHAGO, DOTA REINFORCEMENT LEARNING

  51. RL REQUIRES A LOT OF TRAINING TIME REINFORCEMENT LEARNING

  52. RL REQUIRES A LOT OF TRAINING TIME REINFORCEMENT LEARNING

  53. PRIOR KNOWLEDGE REINFORCEMENT LEARNING

  54. HOW WE CAN ALLOW OUT A.I SYSTEM MAKE TO USE

    PRIOR KNOWLEDGE? REINFORCEMENT LEARNING https://ubisafe.org/explore/demeanure-clipart-prior-knowledge/

  55. META LEARNING

  56. 2. META-LEARNING META LEARNING Walking the street Get the flag

    Go to the target Hit the ball Clear the game Kick the target Find the red ball
  57. None
  58. None

  59. REINFORCEMENT LEARNING LEARNING DEXTEROUS IN HAND MANIPULATION

  60. REINFORCEMENT LEARNING META-RL IN ROBOT

  61. None

  62. ઁоৈ۞࠙җ૓૞ա־Ҋर঻؍ ղ੉ঠӝ

  63. R.L REINFORCEMENT LEARNING Meta Reinforcement Learning Exploration Reinforcement Learning

  64. None
  65. None
  66. None

  67. https://www.facultyfocus.com/articles/teaching-and-learning/wheres-the-curiosity/

  68. WONSEOK JUNG ߊݺо ݅ചо җ೟੗ Ѥ୷о оࣻ Ѻైӝࢶࣻ ੗زର٣੗੉ց ઁಿ٣੗੉ց

    ੘ҋо Է ఋై੉झ౟ ࣿ૘਍৔ ஠ۨ੉ࢲ झ௢ఠ׮੉ߡ ੘о ಁ࣌٣੗੉ց ೾झ౟ۨ੉ց Үਭо ҃৔ஶࢸఢ౟ ؘ੉ఠࢎ੉঱౭झ౟ ੋҕ૑מোҳਗ Үਭо ੋҕ૑מূ૑פয ۽ࠈੋҕ૑מোҳਗ

  69. https://news.joins.com/article/11522250

  70. https://en.paperblog.com/mcdonald-s-a-glimpse-behind-the-making-off-burgers-502103/

  71. https://medium.com/old-media/why-i-started-reading-a-lot-of-books-8e948aecf5fe

  72. https://m.tmall.com/mblist/de_Abr0pFEQpnxc4QwZUVeDgg==.html

  73. https://thoughtcatalog.com/leigh-wright/2014/12/7-incredible-books-on-exploration-that-will-fuel-your-wanderlust/

  74. https://www.pinterest.ch/pin/506655026820432199/

  75. None

  76. https://www.slideshare.net/MAAgcunningham/jmm-mini4-sydney-opera-house-part-2

  77. None

  78. https://www.facultyfocus.com/articles/teaching-and-learning/wheres-the-curiosity/

  79. https://www.rit.edu/science/sms

  80. Ѥ୷о

  81. ߕ٘ࢤഝ

  82. 3FTDVFTDVCBEJWFS

  83. 2018 QuantNet Rankings of Best Financial Engineering Programs Ranked by

    QuantNet in Dec 2017. The most comprehensive 2018 ranking of best Financial Engineering (MFE), Mathematical Finance programs in North America. The 2018 QuantNet ranking of Financial Engineering/Quantitative Finance masters programs in North America provides detailed information on placement and admission statistics from top programs the region, making it uniquely valuable to the quant finance community at large. The 2018 QuantNet rankings are best positioned to help prospective applicants decide where to apply and enroll in those master quantitative programs. 2018 Financial Engineering Programs Rankings Methodology Share your opinion about the 2018 QuantNet ranking. Rank Program Total score Peer assessment score Employment rate at graduation Employment rate 3 months after graduation Average starting base salary (not including bonus) Average GRE Quant of admitted students Tuition Class Size 1 Baruch College, City University of New York Financial Engineering New York, NY 100 4.1 94% 100% $110,000 169.5 $40,980 (non- resident), $27,675 (resident) 36FT, 4PT 1 University of California, Berkeley Financial Engineering Berkeley, CA 100 4.2 84% 96% $108,286 168 $68,725 68FT 3 Carnegie Mellon University Computational Finance Pittsburgh, PA 98 4.3 68% 85% $100,939 169 $83,100 97FT 3 Columbia University Financial Engineering New York, NY 98 3.9 97% 100% $93,000 169 $69,696 99FT

  84. ࢜۽਍ҾӘૐ

  85. None
  86. None

  87. ഑द࠺तೠ੼ਸ଺ਵ࣑աਃ 

  88. EXPLOITATION REINFORCEMENT LEARNING Exploitation Exploration ?

  89. REINFORCEMENT LEARNING META-RL IN ROBOT

  90. https://post.naver.com/viewer/postView.nhn?volumeNo=11968470&memberNo=8254994

  91. https://www.intuitionology.com/intuition-ever-lead-wrong-path/

  92. https://www.bodpress.com/1007/alphago-beginner-s-guide/

  93. https://www.bodpress.com/1007/alphago-beginner-s-guide/

  94. যځೠݾ಴ܳ׳ࢿೞӝਤ೧ࢲח

  95. Ӓݾ಴ܳ׳ࢿೞӝਤೠৈ۞о૑೐۽ࣁझоઓ੤ೠ׮

  96. tޙઁо৵ߊࢤೡө u

  97. tযڃߑߨਸࢎਊೞݶ੉੹ࠁ׮જই૕ࣻ੓ਸө u

  98. ੉૕ޙী׹ਸೞӝਤ೧ࠁా݆਷ҕࠗоਃҳػ׮

  99. None

  100. https://angel.co/ai-build/jobs

  101. Meta Reinforcement Learning Exploration

  102. ةࢲܳాೠр੽҃೷

  103. ૒੽҃೷

  104. WONSEOK JUNG ߊݺо ݅ചо җ೟੗ Ѥ୷о оࣻ Ѻైӝࢶࣻ ੗زର٣੗੉ց ઁಿ٣੗੉ց

    ੘ҋо Է ఋై੉झ౟ ࣿ૘਍৔ ஠ۨ੉ࢲ झ௢ఠ׮੉ߡ ੘о ಁ࣌٣੗੉ց ೾झ౟ۨ੉ց Үਭо ҃৔ஶࢸఢ౟ ؘ੉ఠࢎ੉঱౭झ౟ ੋҕ૑מোҳਗ Үਭо ੋҕ૑מূ૑פয ۽ࠈੋҕ૑מোҳਗ

  105. WONSEOK JUNG ߊݺо ݅ചо җ೟੗ Ѥ୷о оࣻ Ѻైӝࢶࣻ ੗زର٣੗੉ց ઁಿ٣੗੉ց

    ੘ҋо Է ఋై੉झ౟ ࣿ૘਍৔ ஠ۨ੉ࢲ झ௢ఠ׮੉ߡ ੘о ಁ࣌٣੗੉ց ೾झ౟ۨ੉ց Үਭо ҃৔ஶࢸఢ౟ ؘ੉ఠࢎ੉঱౭झ౟ ੋҕ૑מোҳਗ Үਭо ੋҕ૑מূ૑פয ۽ࠈੋҕ૑מোҳਗ

  106. ҾӘೠѪ੉ࢤӟ׮ݶ 
 Ҋ޹ೞҊ 
 ಽযࠁ۰Ҋ֢۱೧ࠁݶ
 যڄөਃ 

  107. ೞҊर਷ োҳ઱ઁо ੓׮ݶ ־ҳٚ૑ োҳपਸ ٜ݅ ࣻ ੓Ҋ ੤߀য ࠁ੉ח

    োҳपਸ ଺ও׮ݶ ־ҳա োҳपী ଵৈೡ ࣻ ੓ח ݽف ݽৈ ೣԋ োҳೞח ࣁ࢚ী হ؍ ױ ೞա੄ ৌܽ োҳࣗ э੉ೡࣻب੓যਃ 

  108. п੗੄ӡਸ଺ই
 ৈ۞࠙݅੄੉ঠӝٜܳ݅যࠁݶযڄөਃ 

  109. ੷ب҅ࣘ&YQMPSBUJPOೞҊ ب੹ೡ҅ദ੉ীਃ

  110. (JUIVC IUUQTHJUIVCDPNXPOTFPLKVOH 'BDFCPPL IUUQTXXXGBDFCPPLDPNXTKVOH #MPH IUUQTXPOTFPLKVOHHJUIVCJP ਋ܻژࠊਃ