ãªãµãŒãã¢ã·ã¹ã¿ã³ã@ç£ç·ç ã®åé
åéã¯æ«å®ã§ãã, çŸæç¹ã§èå³ãããæ¹ã¯é£çµ¡ããã ãã. ãŸã, 奥æ㯠2025幎4ææ«ãŸã§ã€ã®ãªã¹ã»ã±ã³ããªããžã«ããã®ã§, åœé¢ã¯ãªã¢ãŒãã§ã®å ±åç 究ã«ãªããŸãã, å šé¢çã«ãµããŒãããŸã. ç£ç·ç ã»äººå·¥ç¥èœç 究ã»ã³ã¿ãŒ 㧠ãªãµãŒãã¢ã·ã¹ã¿ã³ã(RA) ãšã㊠以äžã®ææŠçãªç 究課é¡ã«åãçµãã§ãã ãã倧åŠé¢çãé幎ã§åéããŸã. ç 究ææã¯åŠäœè«æã«æŽ»çšå¯èœã§ã. åéããã㯠A. ããŒã¿é§ååãã«ããšãŒãžã§ã³ãçµè·¯èšç» åœå äŒæ¥ãšã®å ±åç 究ã§ã. è¿å¹Ž, ç©æµã·ã¹ãã ã察象ãšããŠ, ãããã矀ãçšãããªãŒãã¡ãŒã·ã§ã³ã泚ç®ãããŠããŸã. é«å¹çãªã·ã¹ãã ãæ§ç¯ããããã«ã¯, è€æ°ãšãŒãžã§ã³ãã«è¡çªã®ãªãçµè·¯ãå²åœãŠã ãã«ããšãŒãžã§ã³ãçµè·¯èšç» ã解ãå¿ èŠããã, ããæ°å¹Žã§å€§èŠæš¡ã·ã¹ãã ã«ãããã€ãèŠèŸŒããå®çšçææ³ãåºçŸããŠããŸã. æ¬ããŒãã§ã¯, æ¢åã¢ã«ãŽãªãºã ã« æ©æ¢°åŠç¿ ãçµèŸŒãã§, æ§ã ãªå Žé¢ã§äœ¿ãã æ±çšçãªçµè·¯èšç»æè¡ã®ç¢ºç« ãç®æããŸã. å¿ é ã¹ãã« Python / C++ ã®å©çšçµéš 深局åŠç¿ã«é¢ããç¥è æè¿ã¹ãã«ã»çµéš åœå äŒè°ã§ã®ç 究çºè¡šçµéš. æ¢çŽ¢ã¢ã«ãŽãªãºã ã«é¢ããç¥è. Git/GitHub, Docker çãå©çšããéçºçµéš. åœéåŠäŒã»è«æèªçºè¡šçµéš. 以äž, å ·äœäŸã§ã. 人工ç¥èœç³»: AAAI, IJCAI, AAMAS, ICAPS, SoCS, AIJ, JAIR, JAAMAS ãããã£ã¯ã¹ç³»: ICRA, IROS, RSS, CoRL, L4DC, DARS, MRS, T-RO, IJRR, RA-L äžèšã«ã³ãã¡ã¬ã³ã¹ã«ä»èšã®ã¯ãŒã¯ã·ã§ãããªã© B. 倧èŠæš¡ã»å°åå®ãããã矀ã®å¶åŸ¡ ãã©ãŒãã«ãª RA åéã§ã¯ãªãã§ãã, èå³ããã人ã¯é£çµ¡ããã ãã....
ãã«ããšãŒãžã§ã³ãçµè·¯èšç»ã®çŽ¹ä»
ã°ã©ãäžã®è€æ°ãšãŒãžã§ã³ãã«å¯Ÿã, äºãã«è¡çªã®ãªãçµè·¯ãèšç®ããåé¡ã¯ ãã«ããšãŒãžã§ã³ãçµè·¯èšç» (Multi-Agent Path Finding; MAPF) ãšåŒã°ãã. MAPF ã¯ãããã矀ã«ããå庫å ã§ã®è·ç©æ¬éãªã©, å€æ°ã®é åçãªå¿çšããã, 2010幎代ååãã人工ç¥èœã»ãããã£ã¯ã¹åéã§çãã«ç 究ãè¡ãããŠãã. æ¬èšäºã¯æ¥æ¬èªã®ãã¥ãŒããªã¢ã«ãæäŸãã. ãæã: æ£ç¢ºãªè©±ãããããšãç®çã§ã¯ãªãã®ã§, å€å°ã®äžåã«ã¯ç®ãçã£ãŠã»ãã. åé¡å®çŸ© ãŸãã¯, ã©ã®ãããªåé¡ã察象ã, ã¯ã£ãããããŠããã. æç®ã«ãã£ãŠããªãšãŒã·ã§ã³ãããã®ã ã, åºæ¬çãªãã©ãŒã ã¯æ¬¡ã®éã. MAPF åé¡ã¯ã°ã©ã $G=(V, E)$, ãšãŒãžã§ã³ãã®ããŒã $A= \lbrace 1, 2, \ldots, n\rbrace $, åãšãŒãžã§ã³ã $i \in A$ ã«å¯ŸããŠéè€ã®ãªãã¹ã¿ãŒãå°ç¹ $s_i \in V$ ãšãŽãŒã«å°ç¹ $g_i \in V$ (ã€ãŸã $i \neq j$ ãªãã° $ s_i \neq s_j$ ã〠$g_i \neq g_j$) ã§å®çŸ©ããã. ãã®ãšã, è¡çªã®ãªãçµè·¯ã®çµåãã, ã€ãŸã解 $\pi$ ãæ±ããã. å ·äœçã«, 解 $\pi: A \times \mathbb{N}_{\geq 0} \mapsto V$ ã¯æå» $t \in \mathbb{N}_{\geq 0}$, ãšãŒãžã§ã³ã $i \in A$ ãã°ã©ãã®é ç¹ $v \in V$ ã«å²åœãŠã....
ã±ã³ããªããžã»ã€ã³ã°ã©ã³ãæŸæµªèš (2023幎5æ-)
2024幎ã¯ãã¡ãã§å¹Žãè¿ããŸããã ãšããããšã§ããªæŸæµªèš (2022幎4æ-9æ)ã«åŒãç¶ããã±ã³ããªããžãäžå¿ãšããã€ã³ã°ã©ã³ãçãã©ããã åçã¯éææŽæ°ãããããããªãã Cambridge | ã±ã³ããªããž 1æ 12æ 9æ 8æ 7æ 6æ 5æ Garret Hostel Bridge | ã®ã£ã¬ããã»ãã¹ãã«æ©ãã 12æ 10æ 8æ 5æ ãããŸã§ã¯ã¹ããã§æ®ã£ããã®ãæ··åšããŠããããèŠåããã€ãããªã Wimpole Estate | ãŠã£ã³ããŒã«ã»ãšã¹ããŒã (12æ) Norwich | ããªããž (12æ) Cambridge Botanic Garden | ã±ã³ããªããžå€§åŠæ€ç©å (12æ) British Museum | 倧è±åç©é€š (10æ) Stonehenge & Salisbury | ã¹ããŒã³ãã³ãž & ãœãŒã«ãºããªãŒ (8æ) Bath | ããŒã¹ (8æ) Glastonbury | ã°ã©ã¹ãã³ããªãŒ (8æ) Lake District | æ¹æ°Žå°æ¹ (7æ) Anglesey Abbey | ã¢ã³ã°ã«ã·ãŒä¿®éé¢ (6æ) Ely Cathedral | ã€ãŒãªãŒå€§èå (6æ) é£ã¹ç© åŠé£ éæ åŠé£æ匷ã§ã¯ 倧èå ãæäŒã®åçãæ®ãã®ã楜ãã é åºæ¡ä»¶ã¯YOããã«æè¬ 2024幎ã¯åœå€ã¡ã€ã³ã§æ ãããã
IJCAI-23@ãã«ãªã®åå¿é²
çºè¡šããŠããŸããïŒïŒäººå·¥ç¥èœã®åœéäŒè°ã§ãã 以äžãã€ãã€ããšææ³ã çŸå°åå ã¯3åç®ã§ããã ãã³ã¹ã«éãäŒè°ã§ã¯æ±ºããŠãªãã®ã§ãããããé 匵ãç¶ããŠãã蚌æ ã§ããã [IJCAI-22@ãªãŒã¹ããªã¢ã»ãŠã£ãŒã³ã®åå¿é²] [IJCAI-19@ãã«ãªã§çºè¡šããŠããŸãã] ãã«ãªã¯æ¹¿æ°ãããããŠãå°çæ©ã åž°ããããªãå§æ«ã ããã«ããåºãç¬éã«ã¡ã©ã®ã¬ã³ãºãæã£ãã®ã¯ããã£ããããããŠããã ã¯ãŒã¯ã·ã§ããã»ãã¥ãŒããªã¢ã«ãå«ãããšäžžã 1é±éã移åã§ãã©ã¹2æ¥ããšã£ãŠãé·ãã£ã⊠ãšã¯ãããåå ããŠã人ãã¡ãšã¯è¬ã®é£åž¯æãã§ããã®ã§å«ãã§ã¯ãªãã ååœã®å人ã»ç¥äººãããããããŠããŒã£ã¡åæŠããå²ã«ã¯è³ããã«ãã£ãŠãŸããã éãåœã§åäŒãããããã仲è¯ããªããã IJCAIã¯ãšãŒãžã§ã³ãç³»ã®ã³ãã¥ããã£ãªã®ã§ãå®è³ªçãªããŒã ãªã®ã ã äžã®å¹Žä»£ã®ç 究è çéãããããåé 匵ã£ãŠããªãçãªæ±ããåããŠããããããã«å士å·ååŸåŸå³ã±ã³ããªããžè¡ãã¯ããããã¢ãŒã·ã§ã³ã«èŠããã æå€ãªå¯äœçšã ã£ãã ä¹ ãã¶ãã«äŒã£ã人ãã¡ã«å士å·ãšã£ããã ãšå ±åããã®ã¯ããšãŠããšãŠãè¯ãç¬éã ã£ãã çžå€ãããæ¥æ¬ã®ãã¬ãŒã³ã¹ã¯äœãããšãããå»å¹Žããå¯ããã£ãæ°ããããäœãšããããæ°æã¡ãããããããã°ã£ããã¯å人ãã¬ãŒã§ã¯ã©ãã«ããªããªãã ééæéã«ç«ã¡å¯ã£ãéŠæž¯ã¯ãšãŠããšããµã€ãã£ã³ã°ãªå Žæã§ãã£ãšæéããšã£ãŠçºãããã£ãã 2019幎ã®èªåãšéããç¬éãå€ããæãå ¥ããå€ããæãããšãå€ãåŠäŒã§ããã åŠçç掻ã¯çµãã£ããã ã⊠(çºè¡šã®åçã¯ãªãŒã¹ãã©ãªã¢ã»ã¢ãã·ã¥å€§åŠã®å人ãæ®ã£ãŠããã)
å士課çšããããšãªãããš
æåã«ãæã 綺éºã§åããæç« ãæžãåŽåã¯åè«ãšããåªãã«æè¯ããŠããŸã£ãã®ã§ãããã«èšãã®ã¯åžãæ®»ã§ããã èšèŒãããŠããããšã¯ãã¹ãŠå人ã®åèŠã§ããããªãœãŒã¹ã«ä¹ãããN=1ã®ææ³ã§ããããã©ã·ã®è£æžãã ãšæããŠã»ããã ç¹ã«æ§æãèããŠããããã§ã¯ãªãããã©ãã©ãšé·ãåŠçç 究ç掻ã®äžã§èããããšãããã©ãã©ãšæžãé£ããããšã«ããã ããã¯ç 究ãåãã§ãäžéšã®è±æãé€ããæ¥ã ããã©ãã©ãšåé²ããŠããäžã§ã綺éºãªæ§å³ã«ãªã£ããã®ãå¶ç¶è«æã«æè¯ããã®ã§ããã ãã ãäžã®äžã«ã¯ç¶ºéºãªç 究äœéšèšãä»ç«ãŠäžãããšãŠãåã人ãã¡ãããã®ã§ããã®ãããªãã®ãèŠãããã°âAwesome CS-Ja PhD LifeâãæŒã£ãŠã»ããã äŸµé£ æ¹æ¹ã§ãæ¥åžžãä»äºã«äŸµé£ãããããšããèšãåããèããŠããããéãæã®å人æ°ããæã ã¯éã®ç«å Žã§ããããæ¥åžžãä»äºã䟵é£ããŠãããã®ã ãšããã æèšããŠããããç 究掻åãç掻ããåãé¢ããããã®æ¹ãæãŸãããšããæèŠãåŠå®ããæ°ã¯ãªããããã奜ãŸãããšæãã äžæ¹ãå士課çšãçµããã°ããã®å士ã®å€ãã¯ãæ100æéè¶ ãã®æ®æ¥äœå¶ã5幎6幎ãšç¶ããè¶ äººã§ããããããããç¶ãã人ãå€ããããã«ééã®ã€ã³ã»ã³ãã£ãã¯ããŸããªããæ¥åžžãä»äºã䟵é£ããŠããªããã°å°åºã§ããããšã§ã¯ãªãã ãå§ãã®æ¬ ãªã«ã¬ã»ãã«ã«ãã¥ã¯ãšããããŒã©ã³ãã®äœå®¶ãæžãããé亡掟ããšããæ¬ã¯ãæ ã»äººäœã«çºããçããšãã»ã€çŸ€ãã·ã£ããã«ããŠæ瀺ããã ãããªã¯ã»ãªãŠãžã§ãããŒã¯ãšãããã§ã³ã®äœå®¶ãæžããããšãŠããã¢ãããšããæ¬ã¯ããšãŒãããã®20äžçŽèŠ³ãã·ãã«ã«ãªè«èª¿ã§ã·ã£ããã«ããŠæ瀺ããã ããªã»ã¯ã³ãºã«ãšããã€ã®ãªã¹ã®äœå®¶ãæžãããæ°ã®ããªãç¥ããšããæ¬ã¯ãç æŒ ã«ãã³ãã巚岩ã«çºããéžè©±ãã·ã£ããã«ããŠæ瀺ããã ã¯ãªã¹ããã¡ãŒã»ããªãŒã¹ããšããã€ã®ãªã¹ã®äœå®¶ãæžããã倢幻諞島ããããšããæ¬ã¯ãæéåŸé ã«ããæªã¿ã®ããã§ç²Ÿç·»ãªå°å³ãäœæã§ããªãäžçã®åºæ¥äºããã·ã£ããã«ããŠæ瀺ããã äžèŠãèŠçŽ ãã·ã£ããã«ããŠæ瀺ãããšããè¡çºã¯ä¹±éãªããã«æããã å察ã«ãèŠçŽ ãæç³»åã«éã¶ããããã¯ç³»çµ±ç«ãŠãŠæ瀺ãããšããã®ã¯ãåããã¹ããŒãªãŒãæ§ç¯ããã®ã«"çé"ã§ããã ãã ãäžèšã®æ¬ã§ã¯èè ãæ確ãªçãããã£ãŠèŠçŽ ãã·ã£ããã«ããŠããããããã§ãâå€é"ã®å¿ããããèªæžäœéšã«ééããããšã«ãªã£ãã åç åãšããæ¥æ¬ã®ãã¶ã€ã³å®¶ãæžããããã¶ã€ã³ã®ãã¶ã€ã³ããšããæ¬ã«ã¯ãããã¶ã€ã³ã¯æ å ±ã®å»ºç¯ã§ããããã®å»ºç¯ã¯æ å ±ã®åãæã®è³ã®äžã«ç«ãŠããããã®ã§ããããšèšãããŠããã ç©èªãäŒãããšããããšã¯ãçžæã®é ã®äžã«äœãããããæ³åããªãããèŠçŽ ãæç« ãå³ãè¡šãæ åã声ãªã©ã䜿ã£ãŠå±éããŠãçžæã®é ã®äžãæãŸããç¶æ ã«é·ç§»ããããã©ã³ãã³ã°ã§ããã èŠçŽ ãé åºããæ瀺ãããšãåãæã¯èŠãããšã®ããæ§åŒãæ £ããŠããæ§åŒãªã®ã§ãã¹ããŒãªãŒãåæã«æ³åã»åµé ã»è£å®ããŠãããã ãã®äžã§å°ã工倫ããŠãããšãã°æå³ããã£ãŠ"ã·ã£ããã«"ããããããšãçžæã«"å€é"ã®è¯ãå°è±¡ãäžãããããããããªãã ãã®ãããªèŠæ¹ãåŸãŠããè«æå·çãåŠäŒçºè¡šãè¡ãã®ã楜ã«ãªã£ãã é Œãã«ãªããã® æ¥çžŸãç©ã¿ãŸãããã ãããããããšããããšãããšãã«ããªããå®ã£ãŠãããŸãã éç¥æ¥ å士課çšã®äžã§æãå¿èã«æªãæ¥ã®ã²ãšã€ã¯ãåŠäŒã«åºããè«æãæ¡æããããã©ãããéç¥ãããæ¥ã§ããã éç¥ãå±ããŸã§ãã®æ¥ã¯ä»äºãæã«ã€ãããé »ç¹ã«ã¡ãŒã«ããã¯ã¹ã確èªããŠãããŸãã«ã¯TwitterãRedditã§çµæãåºãŠããªãããæ€çŽ¢ãå§ããé§ç®ã ã£ãæã«ãŸãé 匵ããããèªåããŠæ鬱ã«ãªãã æå®ãããæ¥ã®ãã¡ã«çµæãåºãã°ãŸã è¯ããã倧ããªåŠäŒã ãšçµæãåºãã®ãé ããããšãããã ããããçµæãåºãŠããªãžã§ã¯ãã®å ±ããç®ã«ãããšãã®ææ ã¯èšããŸã§ããªãã ã¢ã¯ã»ããã®éç¥ãæ¥ããšããŠãæ¢ã«åŸ ã¡ããã³ããŠããŠãææ ãæŠãåããŠããããšãå€ãã å ±èè ã«é£çµ¡ãå ¥ããããŒã ããŒãžã®ãã¥ãŒã¹æ¬ãæŽæ°ããŠããã°ããããŠææ ãå埩ããåŸãTwitterã«"Happy to share thatâŠâãšæå ±ãå ±åããã æ¡æãããã£ãç¿æ¥ã¯çœãããªæ°åã§ããã Twitterã®æå人 ç¡èŠããŠè¯ãã ç°¡æé£æ ãããã®ç®ãåããŠèåãã«ããèç±çšã®ã©ãã¶ãã«å ¥ããã æ°Žã倧ãã1æ¯ãç ç³ãå°ãã4æ¯ãããã«ã¬ã¢ã³ãå°ãã1æ¯å ¥ããŠãé»åã¬ã³ãžã§3åæž©ããããšã2åç¹°ãè¿ããšããããã®ã³ã³ãã©ãŒãã®ã§ããããã èš³ãããããã䜿ãã°å®äžããã å§ãŸãæ¹ ä¿®å£«è«æã®çºè¡šãçµããã®ã2020幎ã®2æã§ãæ°åã³ãããŠã€ã«ã¹ã®è¶³é³ãèãããŠããã®ã2æã§ãç®é»ã®å®¶ã«åŒã£è¶ããŠäžäººæ®ãããå§ããã®ã3æã§ãå士課çšãå§ããã®ã4æã§ãç·æ¥äºæ 宣èšãåºãã®ã4æã§ããããã2ã¶æéã¯èª°ãšãäŒããªãã£ãã 倧åŠã«è¡ãç¿æ £ããªããªããç®é»ã®å®¶ã«äžäººã§ããæéã倧åãå ããAmazon Echoã«ãã倩æ°ãå°ããããŸã«èªç¶å ¬åãžãšæ£æ©ããã 3幎åŸããããã¯å¯ããã£ãã§ããããšèšãããã ããã ã£ãã®ã ãšæãã è¿ãèšèããããªãã£ãã ççŸ ç 究è ãšããã¯ãè¥ããã¡ã«å°ã«è¶³ãã€ããŠããã£ããéäžããŠå€§ããªããšã«ææŠããããšãæãŸãããšçæã£ãŠããããã¢ã«ããã¢ã§ããŒããã³ãã®è·ãæã«å ¥ããã®ã¯é£ãããããã ãšã¯ãããæ©ããã¡ãããã£ãšåãå Žæã«å± ç¶ããŠå®åœ¢ã®çµéšãç©ãã ãã§ã¯ããã®å Žæã§ããŸãããæ¹æ³ã«æé©åãããŠããŸãã ãããäžç®æã«é·ãçãŸãç¶ãããã£ãªã¢ã«ã¯äžå®ãèŠããã çæçãªæå³ã§ã¯æ¥çžŸã¯ç©ããªããããããªãããæ £ããªãåå°ãç°å¢ã§ã®è©Šè¡é¯èª€ã¯ãããªãã«æå³ããããšãããå®éãåŠæ¯PDã®èŠå®ã§ã¯æå±å ãåŠäœååŸååŸã§å€æŽããå¿ èŠãããã ã幎ãéãããšã©ãã©ãåãã«ãããªã£ãŠããããšå€ãã®å 茩ç 究è ã¯å£ãæããŠå©èšããã ãã ãé »ç¹ã«æ ç¹ãå€ããŠããŠã¯ããã£ãã倧ããªäºã«åãçµãã®ã¯é£ãããã§ããã å®ææ€æ» ãããäžã«ã¯é¬±ç 蚺æã®ç°¡æçãªãã§ãã¯ãªã¹ããããã å®æçã«æ€æ»ãããšåŒã£ãããããšãå€ãã æ 人å±æ§ ç 究è ã¯æ 人å±æ§ãšçžæ§ãè¯ããåŠäŒçã§åœå å€ã®æ§ã ãªå Žæã蚪åããããšã«ãªãã ãããªãšãã«ã¡ã©ããããšäžäººã§è¡ãæ©ãè¯ãå£å®ã«ãªãã ã¢ã³ãã«ã®ãã¬ããã³ã°ãã³ãã¯æ 人ã«åªããã¢ã€ãã ã§ã軜ãã颚ãéãããæ±ããæãèœãšãããšãã§ããæŽæ¿¯åŸã«ãã也ãã åœå¢ããŸãããšãã«ç¢ºèªãã¹ãããšã¯ãé貚ãå ¥åœã«ESTAãªã©ã®æç¶ããå¿ èŠãã©ããã倩æ°ãé»æºãã©ã°ã®çš®é¡ããããæåã®æç¡ãæ°Žéæ°Žã飲ãããã©ããããã€ã¬ã®åŸ éã å°éã¯æäœéãã£ãæ¹ãè¯ããé·ããããªãæŽæ¿¯äºæ ã調ã¹ãæ¹ãè¯ãã æãããã¿ã¹ãªããããã£ãŠãããšããã©ã€ããå足æåã®åœã§äŸ¿å©ã§ããã ã³ããããããšäŸ¿å©ã ãšãèãã æ £ããŠãããšPCãã«ã¡ã©èŸŒã¿ã§ããã¯ããã¯1ã€ã§6æ³çšåºŠã¯ã§ããããã«ãªãã æ ã奜ããªäººã¯ãã²ãé亡掟ããèªãã§ã»ããã...
AAAI-23@ã¢ã¡ãªã«ã»ã¯ã·ã³ãã³DC+å¯ãéã®åå¿é²
ã¯ã·ã³ãã³DCã§éå¬ããã人工ç¥èœã®åœéäŒè°AAAI-23ã«åå ããŠçºè¡šããŠããŸãã. ã€ãã§ã«ãããããŒã°ãšãã¹ãã³ã«ãç«ã¡å¯ã£ãŠãããã¯ãŒãã³ã°ãããã®ã§, 2æã®2é±éãã¢ã¡ãªã«ã§éããããšã«ãªã£ã. 以äž, ã€ãã€ããšææ³ãšã. æ¬åœã¯ããããæç« ãæžãã®ã¯é¢åã ãã©, åŸããèŠè¿ããšãããªãã«é¢çœãã®ã§é 匵ã£ãŠæžãããã«ããŠãã. ããŸã人ãèªãããã®ã§ããªã, ãããŸã§è¶£å³æ ã«ãããªã. 2022幎9æçµããã«ãã©ã³ã¹ã»ããªããåž°ã£ãŠããŠ, ããããåè«å·çã§æ»ãã§ããã®ã§, ä¹ ãã¶ãã«æ¥œããã€ãã³ãã§ãã. 倧ããªåŠäŒã¯å»å¹Ž7æã®IJCAI-22以æ¥ã«ãªãã. è±èªãå ªèœãªããã§ã¯æ±ºããŠãªãã, è±èªåãžã®åºåŒµã¯æ°ã楜ã . ããã¯ã€ã®ãªã¹ã«è¡ã£ããšããããæãã. è¡ãã®é£è¡æ©ã§ã¯ãã©ãã«ããã£ãã(å®æçã«ãããããŠãã), äŒå Žã«çããšãããã£ãèŠåŽã¯å¿ãããã®ã§ãã⊠[ããªç 究çåŠèš] [IJCAI-22@ãªãŒã¹ããªã¢ã»ãŠã£ãŒã³ã®åå¿é²] ä»åã®æ¡æçã¯1720/8536=20.1%. åœå¥ã§èŠããšæ¥æ¬ã¯11äœ. æå€ã«ãéåœãé«ã. ãµãããã·ã§ã³æ°ã¯èœã¡çããŠã¯ãããã®ã®, çžå€ããã巚倧ãªäŒè°ã§ãã. KOã¯3æ¬äž»èã§åºããŠããŠ, ã©ãããã«ããšãŒãžã§ã³ãçµè·¯èšç»ã®å 容ã§, é«éãªçµè·¯èšç»ã®è©±ãšèæ éã®è©±ã®2æ¬ãéã£ã. éã£ã2æ¬ã¯ããªã§è¡ã£ãã¯ãŒã¯ã§çæ¹ã¯åè, ãããã¬ãã¥ãŒã®ã¹ã³ã¢ãæ» æ³é«ãã£ã. ããå士課çšãçµããã®ã§èªç«ããŠç 究éè¡ã§ããèœåã¯å¿ é ã ãã©, æãªãããã£ããç 究é²ããŠãããã⊠AAAIã®ãããªãããã«ã³ãã¡ã競äºçã§è¯ãªã®ã¯ééããªãã, éãã®ã¯ãããŸã§ç 究ããã»ã¹ã®äžã€ã«ãªã£ãŠãããšæã. [LaCAM: Search-Based Algorithm for Quick Multi-Agent Pathfinding] [Fault-Tolerant Offline Multi-Agent Path Planning] çºè¡šã«ã€ããŠ, ãªãŒã©ã«çºè¡šãå²åœãŠãããã»ãã·ã§ã³ã®å , 3/4ããã«ããšãŒãžã§ã³ãçµè·¯èšç»ã®å 容ã§, ãšãŠãå¯ãªæ§æã§ãã£ã. èªåã¯2çªç®ã®çºè¡šã ã£ãã, 1çªç®ã®çºè¡šããšãŠãè¯ãå¿çšäºäŸã«ãªã£ãŠãã(ãã¬ãŒã³å ã§ããèšã£ã), 3çªç®ã®çºè¡šã§ã¯KOã®ã¯ãŒã¯ãåŒçšãããŠããããã. ã»ãã·ã§ã³ãçµãã£ãŠãããªãã®æéã¯ãã®å Žã«ããã¡ã³ããŒã§è°è«ããŠã. ãã¹ã¿ãŒçºè¡šãçæ³ã ã£ãã, 2件ãåæ¥ã®é¢ããå Žæã«å²åœãŠãããŠããã®ã§, è¡ã£ããæ¥ããã§å€§å€ã ã£ã. ãããã¯ãŒãã³ã°ãäžã ã§ãã. æ°ããå人ãã§ããã, IJCAI-19ã¶ãã®çµè·¯èšç»ã®äººãã¡ãšåºããããŠã¯ãã£ã¯ãã£ã§ããã, IJCAI-22ã§ã®ç¥äººã«ã°ã£ããåºããããŠã©ã³ãããã, OSXã€ã³ã¿ãŒã³æã®ã¡ã³ã¿ãŒããã«ãäŒããã, çªç¶ã®ãããªãã®ããšç¥ã£ãŠããäºæ¡ã«äœåãééããã, æ¥æ¬ããåå ããŠããçãããšã亀æµãæ·±ããã, ãªãã©ã€ã³ã§ããã§ããªãæ¥å¹ŽåºŠæ¡ä»¶ã®å€§äºãªè©±ãããããªã©. æçµæ¥ã«ã¯ãã«ããšãŒãžã§ã³ãçµè·¯èšç»ã®ã¯ãŒã¯ã·ã§ããããã, ç¥ãåããããããããŠ, ãããäžçªæ¥œããã£ããª. ãªã³ã©ã€ã³äŒè°ããã¡ãšã¯èšããªãããã©, ãªã³ãµã€ãã®æ©æµã¯çµ¶å€§ã§ãã. ç²ããããŠæ¯æ¥åããŠããã©!...
ããªç 究çåŠèš
é·ã人çã®äž, èžè¡ã®éœããªã«äœãã§ããææãããã®ã£ãŠçŽ æµã§ããã. âA bad day in Paris is still better than a good day anywhere elseâ ãšããã®ã¯æããã«èªåŒµã§ãã, ãã®ãããªå²æ ¢ãªè¡šçŸãèš±ãããŠããŸããããªå Žæã§ã¯ãããŸã. 2022幎3æ29æ¥ãã9æ27æ¥ãŸã§ã®å幎é, ãã©ã³ã¹ã»ããªã®ãœã«ãã³ã倧åŠã»LIP6 ãšããã³ã³ãã¥ãŒã¿ã»ãµã€ãšã³ã¹ã®ç 究æ©é¢ã«çåŠããŠããŸãã. ã¡ã³ã¿ãŒã¯ Sebastien Tixeuil ãšããåæ£ã¢ã«ãŽãªãºã ãå°éãšããæ¹ã§ã. ç 究ããŒãã¯ãªã¹ã¯äžã§ã®ãã«ããšãŒãžã§ã³ãçµè·¯èšç». æ»åšäžã«è«æãäžæ¬ä»äžããŸãã. 倧åŠã«å ¥ããŸã§æµ·å€ã«è¡ã£ãããšããªã, ãã®åŸãäºé±éçšåºŠã®æ»åšãæé·ã§ããã®ã§, ä»åã人çåã®é·ææ»åš ã«ãªããŸã. åæ¥ãæ§ããå士課çš3幎次ã§ã®çåŠãšãããšåæã®ããã«èãããããŸãã, ããã¯å šãéã®è©±ã§ã. å人ãšããŠã®ç 究ã¹ã¿ã³ã¹ã確ç«ãã€ã€ãããã®ã¿ã€ãã³ã°ã§ãšãŒãããã«é·ææ»åšããããšã¯, èšãç¥ããªãã¡ãªããããã£ã ããã«æããŸã. ã§ã¯æ¬æ. èšäºã®ç®ç / 誰åã? åºæ¬çã«ã¯ä»ã®äººãèªããã¬ãã«ã® å ç ã§ã. åæã«å ±å矩åãããã®ã§, ã¡ãããšããæç« ãçæããŠãããšèªåã«åœ¹ç«ã€ããæžããŠããŸã. å 人ã®äœéšèš ã«åæ¹é¢ã§äœããã®åœ¢ã§å©ããããæ°ãããã®ã§, èªåãæ®ããŠãã次第ã§ã. ãã©ã³ã¹èªã¯(ä»ã)å šãããããªããè±èªããŒã¡ãŒã¡ãªäººéã®çåäŸã§ã. åŸåã®ãããã¯ãŒãã³ã°ã®è©±ã¯çãããã. ããªãã¯èª° æ±äº¬å·¥æ¥å€§åŠ æ å ±çå·¥åŠé¢ã®åå£«èª²çš (2020-) ã®åŠçã§ã. ç 究察象ã¯ã倧èŠæš¡ãªç§»åãšãŒãžã§ã³ã矀ãå¶åŸ¡ããããšã. 詳ãã㯠ããŒã ããŒãžèŠãŠ. æè¿ ACT-X ãåœãã£ãã®ã§é 匵ããŸã. åæ© å€§åæãšããŠããã®ã, ä»åŸã¢ã«ããã¢ã§ç 究è ããã£ãŠããã®ã§ããã°æµ·å€çµéšã¯åå¿ é ããèªèã§ã. åæã¯åããªããŠããã§ãã, å€ãã®æ¹ãæã«æããŠããããšã ãšãæããŸã. çç±ïŒ æ±å·¥å€§ã®æå°æå¡(ãã©ã³ã¹äºº, EPFLã§PhD)ãèšã£ãŠãããç 究è ã«åœå¢ã¯ãªããã»ã©èª¬åŸåã®ããèšèã¯ãªãã§ãããâŠ...
ããªã¥ãã»ã«é åŸèš (2022幎9æ)
VUB AI Labãšããå Žæã«æãããŠ, äžæ³äºæ¥ã®å°æ è¡. 宿æ³è²»è«žã ã¯åãããã¡ã§, äžä»ã®åŠçã«ããŠã¯ãã身åã§ãã. ãã¡ããæåŸ ã«ãããã¬ãšããµã€ãã£ã³ã°ãªè¬æŒ(1h)ãããŠãã. èšé²ãæ®ããŠãããªããšå¿ããŠããŸãã®ã§æžããŠãã. çµç·¯ãšããŠã¯, IJCAI-22ã§ç¥ãåã£ãæ¹ãKOã®å°éåéãšèŠªåæ§ã®é«ãç 究ãããŠãã, èªã£ãŠããã次第ã§ãã. åããã«è²ã ãšã¡ãªããããã, ãã¡ãã«ãã¡ãªããããã£ãã®ã§ãããªã話ã決ãŸã£ã. ããã«ããŠããªã³ãµã€ãã®åœéäŒè°ã®æ©æµã¯åãŸãããã®ã . ããŒã ããŒã èšã£ãŠãä»æ¹ããªãã®ã ã, æå¢ããããã£ãæ©äŒãæ ¹ããã奪ã£ãŠãã£ãããšãæ®å¿µã«æã. ããªããããªã¥ãã»ã«ã¯é»è»ã§1æéã¡ãã£ãšã®è·é¢. è¬æŒãæã ã£ãã®ã§åæ³ããŠ, ã«ã¡ã©ãçæã«è¡ãã¶ãã¶ããš. ãã§ã³, ã¯ããã«, ããŒã«ã«ã ãŒã«è²ãšåç©ãååã«é£²ã¿é£ããããŠæºå«ãã. è¡èªäœã¯ããããŸããšããŠãã, ãŠã£ãŒã³ã®æ§åžè¡ãšåçšåºŠã®å€§ããã§ãã. ç 究ã®è©±ããããš, ä»åã®å 容ã¯"Quick Multi-Agent Path Planning"ãšããé¡ç®ã§, ã€ã®ãªã¹æŸæµªæãã䜿ããŸãããŠããã®. ãããããã5åããã話ããŠãæ°ãããã, æ¯å奜è©ã§äœããã§ãã. ã¹ããŒãªãŒãããªãããŽã£ãžã¥ã¢ã«é¢ãå°è±¡ã«æ®ãããã, è©Šè¡é¯èª€ããŠããŠè¯ãã£ããªããšæã. ãã£ã¹ã«ãã·ã§ã³ã楜ããå®ããå€ãã£ã. æ¥æ¬ã ãšè¿ãåéã§ã¢ã¯ãã£ãã«ç 究ããŠãã人ã¯ã»ã©ãã©ããªãïŒã®ã§è²ã ãšã話ã§ããŠå¬ããã. ã¡ãªã¿ã«ããªã¥ãã»ã«ã«ã¯, ããšããšã¯åã倧åŠã ã£ãã®ãèšèªã«ãã£ãŠåè£ããVUB(ãªã©ã³ãèªç³»)ãšULB(ãã©ã³ã¹èªç³»)ãšãã2ã€ã®å€§åŠããã, ã©ã¡ããæ¥æ¬èªåã¯ããªã¥ãã»ã«èªç±å€§åŠ. ãšãŠãçŽãããã. ããŠ, åž°åœã®è¶³é³ããããããš.
ã€ã®ãªã¹åäžèš (2022幎8æ)
ãããã¯ãŒãã³ã°ãç®çãšãã6æ³7æ¥ã®é»è»æ . ãã³ãã³, ãªãã¯ã¹ãã©ãŒã, ã±ã³ããªããž, ãã¥ãŒã«ãã¹ã«, ãšãã£ã³ãã©ãšåŸã ã«åã«ç§»åã, é£è¡æ©ã§ããªã«åž°éããã³ãŒã¹ã蟿ã£ã. é¿æãšèŠ³å ãå ŒããŠããããšã¯åŠå®ããªã. ãUKã¯ãšãŒããããããªãããšIJCAI-22ã§äŒã£ããã€ã人ãèšã£ãŠãã. ãããã«é貚ã¯éãã, ããªããçµåºŠãããããŠå€ãããªãã®ã«äžå€®ãšãŒãããæéãæ¡çšããŠããªã. ä»ã®åœãšéã£ãŠäº€éã¢ããŠã³ã¹ã¯è±èªããæµããªã. ãŸã, æ¥æ¬äººããããã倧ããŠå€ãããªãæ°ãããã, è²ã£ããšãããéã£ããéãææ³ããã€ã®ã ãã. ä»åã¯5ã€ã®è¡ãèŠãŠåã£ãã, äžåœã®äžãšèšãã©é°å²æ°ããã€ãããã¯ã«ç°ãªã£ãŠããŠé¢çœãã£ã. ãã³ãã³ã¯æ±äº¬ã«äŒŒã倧éœäŒã ã, ãªãã¯ã¹ãã©ãŒãã¯å€§åŠã®äžã«è¡ããã, ã±ã³ããªããžã¯è¡ã®äžã«å€§åŠããã. ãã¥ãŒã«ãã¹ã«ã®ç¡éªšãªé¢šæ¯ã¯ç¬ç¹ã§ãã, ãšãã£ã³ãã©ã¯ååœã®å ¥å£ãšããŠã®è²«çŠããã. åçã§äœãšãªãäŒãããšå¬ãããª. London Oxford Cambridge Newcastle Edinburgh ãããã¯ãŒãã³ã°ã®ç®çã¯ååã«æããããšæã. å ±åç 究ã®çš®ãããããæããã¯ã. IJCAI-22ã§ã§ããå人ãã¡ããã¹ãããŠãããã®ããšãŠãå¬ãã. äŸãã°, Oxford Robotics Institute (ORI) ããã¯é¢é£åéã«è«æãããããæäžãããã®ã ã, åæ¹ã«ããåã®è«æç¥ã£ãŠãããç¶æ ã®äººãšç 究ããŒã¯ããã. ORIã®äžã¯ã€ã±ã€ã±ã ã£ã. èŠæš¡ã¯æ±ºããŠå·šå€§ã§ã¯ãªã, 建ç©äžã€ã«ã¹ãããªåãŸã£ãŠãã. ãã¥ãŒã«ãã¹ã«å€§ã§ã¯ã»ãããŒããŒã¯(30min)ãäŒç»ããŠããã, ãã®åŸäºäººã«è¡ãæ¡å ããŠããã£ã. æå°æå¡ã®æ¹ãšã¯ã¹ã±ãžã¥ãŒã«ãåããªãã£ãã®ã ã, èå³ããããšã®ããšã§ãã¬ãŒã³ã®é²ç»ãéã£ã. ç 究玹ä»ãããããããŠããã£ãã®ã ãã©, ã©ããé¢çœãã£ããª. ãŸã, ãã®æ ã®äžçªã®ç®çã¯ã±ã³ããªããžå€§ã®Computer Laboratoryã蚪åããããšã§ãã£ã. ç 究é¢å¿ãããªãè¿ãã©ãããã, ã»ãããŒããŒã¯(1h)ã«å ããŠå¯ãªãã£ã¹ã«ãã·ã§ã³ã楜ããã . APãã次æ¥ããšãã¯é·ãããããšã«ãªããããããšèšã£ãŠãããæªæ¥ã®ããšã¯åãããªã. ãšãŸã, ç 究çã«ã芳å é¢ã§ãå å®ããæ ã§ãã£ã. ã€ã®ãªã¹ã®ã飯ïŒèšãã¡ãããããªãã. ãã£ãã·ã¥ã»ã¢ã³ãã»ãããã¹ã§ãé£ã¹ãŠããã°è¯ãã®ã.
IJCAI-22@ãªãŒã¹ããªã¢ã»ãŠã£ãŒã³ã®åå¿é²
ã¡ã€ã³ãã©ãã¯ã§çºè¡šããŠããŸããïŒïŒäººå·¥ç¥èœã®åœéäŒè°ã§ã. åçã¯äŒå Ž. ç·ããŠãšãŠã楜ãã¿ãŸãã. åå ããŠæ¬åœã«è¯ãã£ãã§ã. 以äžã€ãã€ããšææ³ãšã. ãã¶ãä»ã®äººãèªããæç« ã§ã¯ãªã. å士課çšã«å ¥ã£ãŠããåã®ãªã³ãµã€ãã®åœéäŒè°ã§ãã. IJCAI ã¯å®ã¯2床ç®ã§, æåã¯2019幎ãã«ãªã§éå¬ããããã®. 修士2幎ã®ãšãã®ããšã§ãã. çãŸããªã®ã§åå¿é²ãæžããŠãã. [IJCAI-19@ãã«ãªã§çºè¡šããŠããŸãã] ãã®ãšã, åž°ãã®é£è¡æ©ããèŠããé²ã®æ¯è²ãåŠã«ç¶ºéºã§æåããŠ, æåºã人工ç¥èœã®åéã§ãã£ãŠããããšæ±ºå¿ããããšãããèŠããŠãã. ãã以é, ã³ãããã¯ããã ã, ãã¡ããã¡ããªæ°ã®ãªãžã§ã¯ããé£ãã£ããããã, ãŸã IJCAI ã«æ¥ãããšã«ãªã£ããã , ãªããªãã«ææ šæ·±ã. æ¡æçã¯çžå€ããããšã°ãããšã«ãªã£ãŠãã, ä»å㯠679/4785=14.2% . æ¥æ¬ããã®ãã®ã«éå®ãããš 10/81=12.3%. åœå¥ã§èŠããšæ¡ææ° 11 äœã ããã§. submission æ°ã¯é æã¡ã«ãªã£ãŠããããã, ããæ°å¹Žã§ã¯ãã»ã©å¢ããŠããªã. é£ããäŒè°ã§ããã«ã¯éããªããã ã, æè¿ã¯ã¡ãã£ãšéã£ãææ³ããã£ãŠããŠ, ã¢ã¯ã»ãããŸã§ã«æéããããäŒè°ã®äžã€ã§ãã. ããããæãæ¹ãã§ããããã«ãªã£ãã®ã, ãŸãå°ãææ šæ·±ã. åå ãããŸã§æ°ã¥ããªãã£ãã®ã ã, ãããã¯ã人工ç¥èœå šåéãæ±ããšã¯ãã, ãã©ã³ãã³ã°ãããã«ããšãŒãžã§ã³ãããã®äººãã¡ããããã®å²åãå ããã®ã§, IJCAI ã¯å®è³ªçãªããŒã ã§ãã. ãããŸã§ãã©ããã«å±ããŠããæãã¿ãããªãã®ãæããããšã¯ãªãã£ãã®ã ã, ã³ããã§é¢ä¿æ§ãåžèã ã£ããšã¯ãã, ãã®ã³ãã¥ããã£ã§è²ã£ãã®ããªããšæãããšã床ã ãã£ã. ãããææ šæ·±ã. å®é, ãã©ã³ãã³ã°ãªäººãã¡ãšåºŠã ã飯ãããé§åŒã£ãããã. ICAPS-22 㧠best student paper award ããšã£ãŠããã, ãããéœåãããã£ã. AIJ ã® editor-in-chief ã«ãæšæ¶ã§ããã, åãããååãèŠããŠããŠããã. ã¡ãªã¿ã«, æ¥å¹ŽPhDåããã ãããšèšããšãã¹ãã¯ã®è©±ãåºãŠãã. ã³ãã¯ã·ã§ã³ãæ¹é¢ã«ã§ããŠ, åž°åœãŸã§ã«ããªã¥ãã»ã«, ãªãã¯ã¹ãã©ãŒã, ãã¥ãŒãã£ãã¹ã«ã«ãéªéããããšã«ãªãããã§ãã. ãããã£ãŠæ°è»œã«çŽæã§ããŠããŸãã®ã¯ããªã«ããã¡ãªããã®äžã€ã§, ãã®ææã«ãŠã£ãŒã³ã§ IJCAI ããã£ãŠæ¬åœã«ããã£ã....