Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
DRL 組み合わせ最適化
Search
newzy
November 24, 2021
Research
8
90
DRL 組み合わせ最適化
newzy
November 24, 2021
Tweet
Share
Other Decks in Research
See All in Research
2021年度-基盤研究B-研究計画調書
trycycle
PRO
0
270
とあるSREの博士「過程」 / A Certain SRE’s Ph.D. Journey
yuukit
9
4.1k
Mechanistic Interpretability:解釈可能性研究の新たな潮流
koshiro_aoki
1
400
Language Models Are Implicitly Continuous
eumesy
PRO
0
190
AIスパコン「さくらONE」のLLM学習ベンチマークによる性能評価 / SAKURAONE LLM Training Benchmarking
yuukit
0
170
投資戦略202508
pw
0
470
日本語新聞記事を用いた大規模言語モデルの暗記定量化 / LLMC2025
upura
0
160
GPUを利用したStein Particle Filterによる点群6自由度モンテカルロSLAM
takuminakao
0
180
Type Theory as a Formal Basis of Natural Language Semantics
daikimatsuoka
1
280
AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data
satai
1
160
RHO-1: Not All Tokens Are What You Need
sansan_randd
1
170
EOGS: Gaussian Splatting for Efficient Satellite Image Photogrammetry
satai
4
480
Featured
See All Featured
We Have a Design System, Now What?
morganepeng
53
7.8k
I Don’t Have Time: Getting Over the Fear to Launch Your Podcast
jcasabona
33
2.4k
Scaling GitHub
holman
463
140k
Side Projects
sachag
455
43k
Product Roadmaps are Hard
iamctodd
PRO
54
11k
The MySQL Ecosystem @ GitHub 2015
samlambert
251
13k
[RailsConf 2023 Opening Keynote] The Magic of Rails
eileencodes
30
9.6k
Into the Great Unknown - MozCon
thekraken
40
2k
The World Runs on Bad Software
bkeepers
PRO
70
11k
Performance Is Good for Brains [We Love Speed 2024]
tammyeverts
11
1.1k
実際に使うSQLの書き方 徹底解説 / pgcon21j-tutorial
soudai
PRO
185
54k
Embracing the Ebb and Flow
colly
87
4.8k
Transcript
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