0QUVOB
.FFUVQ

෺࣭ͷ݁থߏ଄ղੳʹ͓͚Δ
0QUVOBͷԠ༻ ླ໦༤ଠʢ૯߹ݚڀେֶӃେֶɺߴΤωϧΪʔՃ଎ثݚڀػߏʣ  

wླ໦༤ଠ
ʢYuta SUZUKIʣ IUUQTSFTOBOUHJUIVCJP • ૯߹ݚڀେֶӃେֶɹߴΤωϧΪʔՃ଎ثՊֶݚڀՊɹ5೥Ұ؏੍ത࢜՝ఔ5೥ •@resnant (Twitter, GitHub) w3FTFBSDI
*OUFSFTUɿ"*BTTJTUFEͳ෺࣭ɾࡐྉͷཧղ
wػցֶशΛԠ༻ͨ͠৽ͨͳ෺࣭ܭଌٕज़ͷ։ൃ
w࣮ݧσʔλʹ͓͚ΔσʔλϚΠχϯά
w+45
"$5*ݚڀऀʰࡐྉܭଌσʔλͷϞμϦςΟม׵ʱʢ৘ใͱະདྷ
ྖҬʣ
w$PMMBCPSBUPSTʢܟশུʣ
wඌ࡚Յ඙ɺେ੢ਖ਼ًʢ࢈૯ݚ
ਓ޻஌ೳݚڀηϯλʔʣ
wڇٱ঵޹ɺ୩߹ཽయɺઍ༿௚໵
ΦϜϩϯαΠχοΫΤοΫε
w೔໺ӳҳʢ౷ܭ਺ཧݚڀॴʣ
wখ໺ɺ੪౻ɺӋ߹ʢখ໺ݚڀࣨʣ  About me  

wϒϥοΫϘοΫε࠷దԽΛԠ༻ͯ݁͠থߏ଄ղੳΛࣗಈԽͨ͠࿩
w ʮͳʹͬͯʜ0QUVOBΛ࢖ͬͯ࠷దԽ͚͕ͨͩͩ͠ʁʯ
wݚڀॴͷ3FTFBSDI
)JHIMJHIUʹબग़
w์ࣹޫֶձ
ֶੜൃද৆ʢ೥݄ʣ
wࠓ೔ͷߨԋ͸͜ͷ࿦จˣͷ࿩  ࠓ೔ͷߨԋʛࡐྉݚڀʹ͓͚ΔOptunaͷԠ༻ࣄྫ  Ozaki, Y., Suzuki, Y., Hawai, T., Saito, K., Onishi, M. & Ono, K. Automated crystal structure analysis based on blackbox optimisation. npj Comput. Mater 6, 75 (2020). Խֶ޻ۀ೔ใ
 

• ೔ຊ࠷େͷ෺ཧɾࡐྉͷݚڀॴʢ෺ཧతʹ΋େ͖͍ʣ • Ӊ஦ɾૉཻࢠ͔Βࡐྉɾੜ෺·Ͱ෯޿͍ݚڀ • զʑ͸Xઢʢ์ࣹޫʣΛ࢖ͬͯ෺࣭Λௐ΂ΔݚڀΛ͍ͯ͠·͢ • ์ࣹޫɿ΄΅ޫ଎Ͱӡಈ͢Δిࢠ͔Β์ࣹ͞ΕΔ์ࣹઢ • KEK stands for ߴΤωϧΪʔՃ଎ثݚڀػߏ  ߴΤωϧΪʔՃ଎ثݚڀػߏʢKEKʣ  

w9ઢճંʢ93%ʣ͸݁থߏ଄ղੳͷجຊख๏ͱͯ͠ීٴ
w93%ύλʔϯ͔Β݁থߏ଄͕Θ͔Δʢ೥ϊʔϕϧ৆ʣ
w෺࣭ݚڀɾ࢈ۀΛࢧ͑ΔෆՄܽͳπʔϧͷҰͭ  93%ଌఆ૷ஔ /E࣓ੴͷ93%ύλʔϯ /E࣓ੴͷ݁থߏ଄ ωΦδϜʢ/Eʣ࣓ੴ
ʢ/E 'F #ʣ ϐ ʔΫϑΟοςΟϯά
 * Chen, P. A. et al. J. Magn. Mag. Mater. 370, 45–53 (2014). ݁থߏ଄ղੳɿ෺࣭ɾࡐྉݚڀͷ࢝఺ 

w͸΍Ϳ͞ͷϦϡ΢ά΢ࢼྉղੳͰ΋93%͕׆༂
wඇഁյ͔ͭɺඍখͳࢼྉͰ΋ଌఆ͕Ͱ͖Δ
wࠓि͔Β,&,
1IPUPO'BDUPSZͰ΋ଌఆ͕ਐߦத  ݁থߏ଄ղੳɿ෺࣭ɾࡐྉݚڀͷ࢝఺ ͜Ε͕Ϧϡ΢ά΢ͷ࠭
ʢଟ෼NN͙Β͍ʣ 1IPUPO
'BDUPSZ
#-"ɹʢۃݶ৚݅Լਫ਼ີ୯݁থ9ઢճં#-ʣ $
ߴΤωϧΪʔՃ଎ثݚڀػߏ
෺࣭ߏ଄Պֶݚڀॴ $
ߴΤωϧΪʔՃ଎ثݚڀػߏ
෺࣭ߏ଄Պֶݚڀॴ 

w9ઢճંʢ93%ʣ͸݁থߏ଄ղੳͷجຊख๏ͱͯ͠ීٴ
w93%ύλʔϯ͔Β݁থߏ଄͕Θ͔Δʢ೥ϊʔϕϧ৆ʣ
w෺࣭ݚڀɾ࢈ۀΛࢧ͑ΔෆՄܽͳπʔϧͷҰͭ  93%ଌఆ૷ஔ /E࣓ੴͷ93%ύλʔϯ /E࣓ੴͷ݁থߏ଄ ωΦδϜʢ/Eʣ࣓ੴ
ʢ/E 'F #ʣ ϐ ʔΫϑΟοςΟϯά
 * Chen, P. A. et al. J. Magn. Mag. Mater. 370, 45–53 (2014). w93%ύλʔϯͷղੳʹ͸ɺ෺ཧϞσϧΛ༻͍ͨσʔλղੳ͕ඞཁ
w͔͠͠ɺ͜ͷखؒͱ͕࣌ؒ՝୊ʹɾɾɾ😭 ݁থߏ଄ղੳɿ෺࣭ɾࡐྉݚڀͷ࢝఺ 

 ࣮ࡍͷଌఆͷ༷ࢠ  ͍͋ͪ43ɹߴ଎93%ϏʔϜϥΠϯ
#-4 •ଌఆσʔλͷղੳ͕ݚڀͷϘτϧωοΫʹ
•93%ଌఆɿ݅EBZ
•93%ղੳɿ







݅EBZɹˠσʔλͷେ൒͸ଌ͚ͬͨͩͰະ׆༻ʹʜ 93%ύλʔϯͷྫ 

 XRDύλʔϯղੳͷྲྀΕ • Ϧʔτϕϧτղੳɿ෺ཧϞσϧͷ͋ͯ͸Ίʹجͮ͘ค຤ճંύλʔϯͷղੳ๏ • ॳظ݁থߏ଄͔Βܭࢉ͞ΕΔXRD͕ܭଌXRDʹҰக͢ΔΑ͏ɺߏ଄ΛΞοϓσʔτ͢Δ  1SPT wચ࿅͞Ε࢖͍΍͍͢ख๏
w݁থͷ༷ʑͳ৘ใ͕ಘΒΕΔ $POT wࢼߦࡨޡͰܾΊΔύϥϝʔλଟ਺
wखؒͱ͕͔͔࣌ؒΔʢ݅਺࣌ؒʣ
wύϥϝʔλௐ੔ʹܦݧ͕ඞཁ 

 ݚڀͷ֓ཁ • ϦʔτϕϧτղੳΛϒϥοΫϘοΫε࠷దԽʢBBOʣͷ࿮૊ΈͰࣗಈԽͨ͠ʢBBO-Rietveldʣ • BBOɿ਺ཧ࠷దԽͷ໰୊ઃఆͷҰͭɻؔ਺ܗ΍ޯ഑͕Θ͔Βͳ͍ؔ਺ͷ࠷దԽΛѻ͏ • ػցֶशϞσϧͷϋΠύʔύϥϝʔλ࠷దԽʢHPOʣ͸ɺBBOͷ୅දతͳԠ༻ • ίΞΞΠσΟΞɿϦʔτϕϧτղੳͷࢼߦࡨޡ΋ɺHPOಉ༷ʹղ͚ΔͷͰ͸ʁ  ϦʔτϕϧτղੳͷϫʔΫϑϩʔ 

 ݚڀͷ֓ཁ  f Blackbox function R
Λ࠷খԽ͢ΔઃఆΛ୳͢໰୊ͱͯ͠ղ͚Δ PQUJNJ[BUJPO
MPPQ ϦʔτϕϧτղੳͷϫʔΫϑϩʔ • ϦʔτϕϧτղੳΛϒϥοΫϘοΫε࠷దԽʢBBOʣͷ࿮૊ΈͰࣗಈԽͨ͠ʢBBO-Rietveldʣ • BBOɿ਺ཧ࠷దԽͷ໰୊ઃఆͷҰͭɻؔ਺ܗ΍ޯ഑͕Θ͔Βͳ͍ؔ਺ͷ࠷దԽΛѻ͏ • ػցֶशϞσϧͷϋΠύʔύϥϝʔλ࠷దԽʢHPOʣ͸ɺBBOͷ୅දతͳԠ༻ • ίΞΞΠσΟΞɿϦʔτϕϧτղੳͷࢼߦࡨޡ΋ɺHPOಉ༷ʹղ͚ΔͷͰ͸ʁ 

 ݚڀͷ֓ཁ  • ۩ମతͳϫʔΫϑϩʔ • Ϧʔτϕϧτղੳͦͷ΋ͷ͸طଘιϑτ΢ΣΞʢGSAS-IIʣͰ࣮ߦ • BBOͰ୳ࡧ͢Δͷ͸GSASͷઃఆʢ݁থߏ଄ͷॳظ஋Ͱ͸ͳ͍ɻޙड़ʣ • Rwp ʢXRDύλʔϯͷ࢒ࠩʣΛখ͘͢͞ΔઃఆΛ୳͢ ϒϥοΫϘοΫε࠷దԽ ෺ཧϞσϧઃఆ ॳظύϥϝʔλ Ϧʔτϕϧτਫ਼ີԽ ਫ਼ີԽ͞Εͨ݁থߏ଄ Rwp(x) background function : Chebyshev degree of background function : 9 background refinement : True peak shape refinement : True etc. ϒϥοΫϘοΫε࠷దԽϧʔϓ ઃఆʹج͖ͮɺ Ϧʔ τϕϧ τਫ਼ີԽΛ࣮ߦ Minimise Rwp(x) subject to c(x) ... initial crystal structure instrument parameters etc. ... Trial 1 Trial 2 Trial 3 ͜Ε·Ͱͷਫ਼ີԽ݁ՌΛجʹ ৽͍͠ઃఆxΛੜ੒ ੍໿৚݅Λຬ͔ͨ֬͢ೝ - - 

 ख๏ʛΞϧΰϦζϜɾ୳ࡧۭؒʹ͍ͭͯ  • ##0ͷΞϧΰϦζϜʹ͸5SFFTUSVDUVSFE
1BS[FO
&TUJNBUPSʢ51&ʣΛ༻͍ͨ
• ϕΠζ࠷దԽͷҰछ
• ͭͷࡐྉΛ༻͍ͯݕূͨ͠ʢલऀͭʹ͍ͭͯ঺հʣ
• :0
%4.0
%Z4S.O0
-J$P0
• ࠷దԽͨ͠(4"4ͷઃఆม਺Ұཡʢ࿈ଓɺΧςΰϦΧϧɺίϯσΟγϣφϧม਺͕ࠞࡏʣ 

 ख๏ʛ࣮૷ͷҰ෦Λ঺հ  • ίʔυΑΓҰ෦ൈਮ
IUUQTHJUIVCDPNRVBOUVNCFBN##03JFUWFME
• ֤ม਺ͷϑΟοςΟϯάॱ͸యܕతͳॱ൪ʹैͬͨ
࠷ॳʹண໨͢ΔUIFUBྖҬΛࢦఆ όοΫάϥ΢ϯυؔ਺Λઃఆ
ʢؔ਺ܗɾ࣍਺ʣ όοΫάϥ΢ϯυΛϑΟοςΟϯά ʢଓ͘ʣ ໨తؔ਺ɿ(4"4**ΛɺύϥϝʔλΛೖΕΔͱ3XQΛฦؔ͢਺ͱΈͳ͢ 

w ιϑτ΢ΣΞͷग़དྷ͕ྑ͔ͬͨ
w ࢖͍΍͍͢"1*
w ୳ࡧۭؒͷఆ͕ٛॻ͖΍͍͢
w EFpOFCZSVOελΠϧ͸΍͸Γศར
w Ϛϧνϊʔυ΁ͷεέʔϧ͕༰қ
w εύίϯʢϊʔυɺ෺ཧ$16ʣ্Ͱͷ
Քಇ࣮੷͋Γ
w ৴པͰ͖Δ։ൃମ੍
w ܧଓత͔ͭ׆ൃͳ։ൃ͕ݟࠐ·Εͨ
w ՄࢹԽͳͲ݁Ռͷ෼ੳػೳ΋ॆ࣮ͭͭ͋͠Δ
w ࠃ࢈ιϑτ΢ΣΞΛԠԉ͍ͨ͠ OptunaΛ࠾༻͢Δʹࢸͬͨཧ༝  

w 3%#ͷύϑΥʔϚϯε͕໰୊ʹͳͬͨʢղܾࡁʣ
w ॳظͷ0QUVOBͰ͸3%#͔ΒͷಡΈग़͕͠஗͘ɺ
ඇৗʹେ͖͍O@USJBMʢສͱ͔ʣͰTUVEZΛ࣮ߦ͢ΔͱɺͦͷTUVEZΛMPBEͰ͖ͳ͔ͬͨ
w 0QUVOB
͋ͨΓͷόʔδϣϯΞοϓͰղܾ͞Εͨ
w O@KPCTҾ਺Λେ͖͘͢Δͱ3%#͕σουϩοΫ͢Δʢղܾࡁʣ
w 3%#΁ͷίωΫγϣϯϓʔϧͷσϑΥϧτ͕ʢ͔֬ʣ
w O@KPCTΛΑΓେ͖͘͢Δ৔߹ɺίωΫγϣϯϓʔϧΛେ͖͘͢Δඞཁ͋Γ
• optuna.storages.RDBStorage(url, engine_kwargs={“pool_size”: 128}) w O@KPCT͸ݱࡏEFQSFDBUF͞Εͨʢ͕෮׆͢Δ͔΋͠Εͳ͍Β͍͠ʣ
w (4"4**ʢϦʔτϕϧτղੳιϑτʣͷṖόάΛ౿Ή
w ਓ͕ؒ(6*Ͱૢ࡞͢ΔલఏͰฒྻԽ͸૝ఆ͞Ε͍ͯͳ͍
w PCKFDUJWFͷதͰ(4"4ΛJNQPSUͯ͠ରԠ ࣮૷தʹۤ࿑ͨ͠఺ͳͲ  

Demo  ʢˡಈըʣ 

(a) (b) • 30෼ఔ౓Ͱख़࿅ऀΑΓྑ͍͋ͯ͸Ί݁ՌΛಘͨ • best Rwp = 6.610% • 6.96% by human expert, 7.23% by Auto FP • AutoFPɿΤΩεύʔτγεςϜʢϧʔϧͷ૊Έ߹Θͤʣʹجͮ͘طଘख๏ 5IF
PQUJNJTBUJPO
IJTUPSZ
PG
3XQ
3JFUWFME
QMPU
PG
UIF
CFTU
DPOpHVSBUJPO BBO-Rietveldͷ݁ՌʛY2O3  

(a) (b) 5IF
PQUJNJTBUJPO
IJTUPSZ
PG
3XQ
3JFUWFME
QMPU
PG
UIF
CFTU
DPOpHVSBUJPO BBO-Rietveldͷ݁ՌʛDSMO • 30෼ఔ౓Ͱख़࿅ऀΑΓྑ͍͋ͯ͸Ί݁ՌΛಘͨ • best Rwp = 9.211% • 9.76% by human expert, 12.8% by Auto FP • AutoFPɿΤΩεύʔτγεςϜʢϧʔϧͷ૊Έ߹Θͤʣʹجͮ͘طଘख๏  

• ཚ਺Λม͑ͯ100ճࢼߦ͠ɺ୳ࡧͷ҆ఆੑΛௐ΂ͨ • Y2O3 Ͱ90%ɺDSMOͰ99%͸ख़࿅ऀΑΓྑ͍͋ͯ͸Ί͕ಘΒΕͨ • ʮRwp ͕খ͍͞ʯ= ྑ͍͋ͯ͸Ίͱఆٛ • طଘख๏ʢAutoFPʣ͸ط஌෺࣭ʹΦʔόʔϑΟοτ͍ͯ͠Δʁ ݁Ռʛ୳ࡧͷ҆ఆੑධՁ ճͷࢼߦʹΑΓಘΒΕͨ3XQͷώετάϥϜ :0 "VUP'1͸%4.0Ͱ͸ඍົ
3XQ


 ճͷࢼߦʹΑΓಘΒΕͨ3XQͷώετάϥϜ %4.0  

w ఏҊख๏ʹΑΓɺ΄΅ಉ͡
3XQ͕ͩߏ଄͕ҟͳΔީิ͕ݟ͔ͭͬͨ
w ख़࿅ऀͱ΄΅ಉ͡ߏ଄
w શ͘ҟͳΔߏ଄΋ݟ͚ͭͨ Best by BBO (Rwp = 9.211%) Similar to human expert by BBO (Rwp = 9.577%) Best by human expert (Rwp = 9.775%) Another interesting candidate by BBO (Rwp =9.484%) CZ
IVNBO
FYQFSU
3XQ PVUMJFS
CZ
##0
3XQ ݁ՌʛಘΒΕͨ݁থߏ଄ͷ෼ੳ .O0ͷ݁߹͕֯ҟͳΔ ֤఺͕ͭͷ݁থߏ଄ʹରԠɺڑ཭͸ߏ଄ͷྨࣅ౓Λද͢  

 αϚϦʔ  w ܭଌσʔλ΁ͷ෺ཧϞσϧ͋ͯ͸ΊΛ##0ͷ໰୊ͱͯ͠ఆࣜԽͨ͠
• BBO-Rietveld ʢhttps://github.com/quantumbeam/BBO-Rietveldʣ w ͋ͯ͸ΊΛվળ͢ΔͨΊͷࢼߦࡨޡΛࣗಈԽͰ͖ͨ
w ख़࿅ऀͷ݁Ռͱಉ౳ͷ݁থߏ଄ɾΑΓྑ͍͋ͯ͸·ΓΛಘͨ
w ܭࢉ࣌ؒ͸ϊʔυͰ෼ఔ౓ʢฒྻԽ΋Մೳʣ
• ҟ෼໺ͷࢹ఺Ͱࡐྉͷ໰୊Λଊ͑௚͢ͱγϯϓϧʹղ͚ͨ
• ֶࡍݚڀͷ͓΋͠Ζ͞
• ܭࢉػͱΞϧΰϦζϜͰղܾՄೳͳࡐྉͷ໰୊͸·ͩ·ͩ͋Δ
• ʰਓؒɾ"*ɾϩϘοτ͕ڠಇ͢Δࡐྉ։ൃʱΛ໨ࢦ͍ͯ͠·͢
• ΫϥΠϛϯάߦ͖·͠ΐ͏ HSBEJFOU
BTDFOU͢Δླ໦ˠ