Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Sign up for free
Menu
Search
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Pricing
Search
Sign in
Sign up for free
DRL 組み合わせ最適化
Search
newzy
November 24, 2021
Research
120
8
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
DRL 組み合わせ最適化
newzy
November 24, 2021
Transcript
26 slides
POMO: Policy Optimization with Multiple Optima for Reinforcement Learning Kwon,
Yeong-Dae, et al. NeurIPS, 2020, vol.33
ཁ •Έ߹Θͤ࠷దԽʹ͓͚ΔɼਂڧԽֶश ͰͷFOEUPFOEͷۙࣅղ๏ɽ •طଘͷਂڧԽֶशख๏ͱൺֱͯ͠ɼ ܭࢉ࣌ؒɾਫ਼ͱʹେ͖͘վળͨ͠ •८ճηʔϧεϚϯͳͲͰݕূɽ 2/26
ಋೖ
Έ߹Θͤ࠷దԽ •८ճηʔϧεϚϯૹܭըɼφοϓβοΫ ͳͲʹද͞ΕΔΑ͏ͳ࠷దͳΈ߹ΘͤΛٻΊΔɽ 4/26 精度 計算時間 厳密解法 最適 遅い 近似解法
最適に 近い 早い https://onl.tw/vzkASMX
ڧԽֶशʢ3FJOGPSDFNFOU-FBSOJOH3-ʣ •3-ɿஞ࣍తͳҙࢥܾఆΛղ͘ख๏ɽ ྦྷੵใु͕࠷େʹͳΔΑ͏ͳํࡦΛݟ͚ͭΔ͜ͱ͕తɽ 5/26 ઃఆͱͯ͠ɼঢ়ଶू߹ɼߦಈू߹ɼใुؔΛ ઃఆ͢Δඞཁ͕͋Δɽ https://onl.tw/98fQVvW
ํࡦϕʔεͷ3&*/'03$& 6/26 •ํࡦ 𝜋 𝑠 ɿঢ়ଶ𝑠ʹ͓͚Δߦಈ𝑎Λग़ྗ͢Δؔ •𝜋! ɿύϥϝʔλ 𝜃ͰύϥϝʔλԽ͞Εͨํࡦ •ํࡦͷߋ৽ࣜɿ𝛼ֶशɼ𝐽
𝜋! తؔ 𝜃 ← 𝜃 + 𝛼∇! 𝐽 𝜋! •ํࡦޯͷࣜɿ𝔼ظɼ𝑅" ऩӹɼ𝑏 𝑠 ϕʔεϥΠϯ ∇! 𝐽 𝜋! = 𝔼#! ∇! log 𝜋! ⋅ 𝑅" − 𝑏 𝑠
ઌߦݚڀ
1PJOUFS/FUXPSLTʢʣ Έ߹Θͤ࠷దԽͰར༻͢ΔωοτϫʔΫ •ॏෳͳ͘બ͠ɼग़ྗύλʔϯྻΛੜ͢Δɽ •ೖྗใ͔Βಛநग़Λߦ͏FODPEFSͱɼFODPEFS ͷग़ྗΛར༻ͯ͑͠ͱͳΔܦ࿏Λग़ྗ͢ΔEFDPEFS͔ ΒͳΔɽ •FODPEFSͱEFDPEFSʹ-45.Λ༻ɽ 8/26
"UUFOUJPO .PEFMʢʣ 1PJOUFS/FUXPSLTͷվྑ൛ •1PJOUFS/FUXPSLTಉ༷ɼ&ODPEFSͱ%FDPEFSΛ༻͢Δ Ϟσϧɽ •-45.ഇࢭ͠ɼ.VMUJIFBE"UUFOUJPOΛ࠾༻ɽ 9/26
ख๏
ຊจͷख๏ͷΞΠσΞ 11/26 ࠷ॳͷߦಈɼޙͷΤʔδΣϯτͷߦಈʹେ͖͘ӨڹΛ༩͑Δɽ Έ߹Θͤ࠷దԽʹΑ͘ݟΒΕΔରশੑΛར༻ɽ
10.0 •3&*/'03$&XJUI#BTFMJOFɿయܕతͳํࡦޯϕʔεͷ 3-ΞϧΰϦζϜΛ༻ɽ •ෳͷҟͳΔ։࢝ߦಈΛࢦఆ͠ɼෳͷߦಈܥྻʢيಓʣ ΛಘΔɽ •ʻ45"35ʼτʔΫϯΛ༻͍ͳ͍ɽ 12/26 従来 POMO
10.0 ∇! 𝐽 𝜃 ≈ 1 𝑁 6 $%& '
𝑅 𝜏$ − 𝑏$ 𝑠 ∇! log 𝑝! 𝜏$ ∣ 𝑠 𝑤ℎ𝑒𝑟𝑒 𝑝! 𝝉$ ∣ 𝑠 ≡ @ "%( ) 𝑝! 𝑎" $ ∣ 𝑠, 𝑎&:"+& $ يಓ 𝝉$ = 𝑎& $ , 𝑎( $ , … , 𝑎) $ GPS 𝑖 = 1,2, … , 𝑁 ڞ༗ϕʔεϥΠϯ 𝑏$(𝑠) = 𝑏TIBSFE (𝑠) = 1 𝑁 6 ,%& ' 𝑅 𝝉, GPS 𝑖 = 1,2, … , 𝑁 13/26
܇࿅෦ͷٖࣅίʔυ 14/26
*OTUBODF"VHNFOUBUJPOɿਪख๏ •ը૾ॲཧͷσʔλΦʔάϝϯςʔγϣϯ͔Βணɽ •ࠓճ͏࠲ඪɼYͷ୯Ґਖ਼ํܗʢୈҰݶʣͷ ͷΛར༻ɽ 15/26 今回使う Instance Augmentation
ਪ෦ͷٖࣅίʔυ 16/26
࣮ݧ
࣮ݧ ࣮ݧ༰ •10.0Λ༻͍ͯɼҎԼͷΛղ͍ͨ݁ՌΛଞͷදతख๏ͱ ൺֱɽ ८ճηʔϧεϚϯ ༰ྔ੍͋Γͷૹܭը φοϓβοΫ
18/26
ֶशۂઢɿ८ճηʔϧεϚϯ 19/26 50地点 100地点
८ճηʔϧεϚϯʢ541ʣ 20/26
८ճηʔϧεϚϯʢ541ʣ 21/26
༰ྔ੍͋Γͷૹܭըʢ$731ʣ 22/26
φοϓβοΫʢ,1ʣ 23/26
࣮ݧͷ·ͱΊ •ҟͳΔઃఆͷͭͷΈ߹Θͤ࠷దԽʹରͯ͠ɼ ಉҰͷ܇࿅ख๏ͱ//ΞʔΩςΫνϟΛ༻͍ͯ༗ͳ݁ՌΛ ಘͨɽ •܇࿅ɾਪख๏ͱͯ͠ͷ10.0ɼਪख๏ͱͯ͠ͷ *OTUBODF"VHNFOUBUJPOͲͪΒޮՌతͳख๏Ͱ͋Δ͜ͱ Λ֬ೝͨ͠ɽ 24/26
·ͱΊ ຊจͰΈ߹Θͤ࠷దԽʹ͓͍ͯɼରশੑΛར༻ ͯ͠3-ͷαϯϓϧޮਫ਼ ਪ࣌ؒΛॖ͢Δख๏Λ հͨ͠ɽ 25/26
ࢀߟจݙ ,XPO :FPOH%BF FUBM10.01PMJDZ0QUJNJ[BUJPOXJUI .VMUJQMF0QUJNBGPS3FJOGPSDFNFOU-FBSOJOH "EWBODFTJO /FVSBM*OGPSNBUJPO1SPDFTTJOH4ZTUFNT
,PPM 8PVUFS )FSLF WBO)PPG BOE.BY8FMMJOH"UUFOUJPO -FBSOUP4PMWF3PVUJOH1SPCMFNT *OUFSOBUJPOBM$POGFSFODF PO-FBSOJOH3FQSFTFOUBUJPOT 7JOZBMT 0SJPM .FJSF 'PSUVOBUP BOE/BWEFFQ+BJUMZ1PJOUFS /FUXPSLT "EWBODFTJO/FVSBM*OGPSNBUJPO1SPDFTTJOH 4ZTUFNT 26/26
Other Decks in Research
See All in Research
モデル評価を効率的にする方法
ftakahashi
0
250
[SNLP2026] Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
wataruuuuu
0
370
J-STAGEの現況と全文XML登載必須化について
xspa2012
0
410
Méthane 2030 : Développer des solutions de réduction des émissions de méthane entérique : 1ers résultats
institutdelelevage
PRO
0
120
Byzantine Eventual Consistency for Reliable Local-First Software
ept
0
410
論文読み会 SNLP2026 Tau2-Bench: Evaluating Conversational Agents in a Dual-Control Environment
s_mizuki_nlp
0
350
Bilan travail en élevage : première restitution des enquêtes 2025-2026
institutdelelevage
PRO
0
210
SPACE 2026 - Renforcer l’attractivité des métiers des productions animales dans les établissements d'enseignement agricole
institutdelelevage
PRO
0
120
Santé et fertilité des sols : ce que nous disent les pratiques dans les systèmes de polyculture-élevage
institutdelelevage
PRO
0
160
La génomique au service de la fromageabilité du lait grâce aux spectres MIR
institutdelelevage
PRO
0
230
生成AI時代の開発方法論と、メールサーバー運用の新しいかたち
hirachan
0
240
Bilan travail en élevage : première restitution des enquêtes 2025-2026
institutdelelevage
PRO
0
300
Featured
See All Featured
Data-driven link building: lessons from a $708K investment (BrightonSEO talk)
szymonslowik
1
1.4k
Building a Scalable Design System with Sketch
lauravandoore
464
34k
Reflections from 52 weeks, 52 projects
jeffersonlam
356
22k
The SEO Collaboration Effect
kristinabergwall1
1
620
Jess Joyce - The Pitfalls of Following Frameworks
techseoconnect
PRO
1
480
Primal Persuasion: How to Engage the Brain for Learning That Lasts
tmiket
0
520
Templates, Plugins, & Blocks: Oh My! Creating the theme that thinks of everything
marktimemedia
31
3k
The Invisible Side of Design
smashingmag
301
52k
Embracing the Ebb and Flow
colly
88
5.2k
Statistics for Hackers
jakevdp
799
230k
Applied NLP in the Age of Generative AI
inesmontani
PRO
4
2.6k
The World Runs on Bad Software
bkeepers
PRO
72
12k