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
人と音の情報学研究室配属説明会−研究紹介
Search
Yuya Yamamoto
October 03, 2020
Research
96
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
人と音の情報学研究室配属説明会−研究紹介
2019年10月に行われた配属説明会の研究紹介です.
Yuya Yamamoto
October 03, 2020
More Decks by Yuya Yamamoto
See All by Yuya Yamamoto
GMI44@Music and Language研究のサーベイ報告
yamathcy
1
190
APSIPA 2023 Toward Leveraging Pre-Trained Self-Supervised Frontends for Automatic Singing Voice Understanding Tasks: Three Case Studies
yamathcy
0
150
muana vol.11 音楽識別の事前学習モデル
yamathcy
1
860
国際会議ISMIR2022報告(山本分)
yamathcy
0
280
Do It Yourself: Sonic Visualiserで音楽分析してみよう
yamathcy
0
2.1k
2022年度情報学学位プログラム説明会 学生体験談
yamathcy
0
210
LSPC博士前期チュートリアル
yamathcy
0
240
MULTIMODAL METRIC LEARNING FOR TAG-BASED MUSIC RETRIEVAL@ICASSP2021読み会
yamathcy
0
1.9k
SIGMUS130-yamamoto
yamathcy
0
90
Other Decks in Research
See All in Research
[webinaire du Ciirpo] Agnelages plein air : des éleveurs partagent leurs pratiques
institutdelelevage
PRO
0
140
Pretrain Where? Investigating How Pretraining Data Diversity Impacts Geospatial Foundation Model Performance
satai
3
180
TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis
satai
3
110
2026年 オープンキャンパス 研究室紹介
junkurihara
0
240
MIRU2026 チュートリアル講演2:三次元データ処理の動向
nnchiba
8
5.3k
Hitting the buffers
beale
0
240
OWASP AISVS - C7
shiell
2
830
IA for theory
gpeyre
1
480
XDPerf: A High-Performance Traffic Generator Built with WASM and eBPF
takehaya
1
340
人間中心の意思決定支援AI
yukinobaba
PRO
7
4.2k
医療LLMの現在地〜最新研究から社会実装までを考える〜
kento1109
1
1.8k
[SNLP2026] Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
wataruuuuu
0
330
Featured
See All Featured
Jess Joyce - The Pitfalls of Following Frameworks
techseoconnect
PRO
1
440
Navigating Algorithm Shifts & AI Overviews - #SMXNext
aleyda
1
1.6k
Redefining SEO in the New Era of Traffic Generation
szymonslowik
1
470
Information Architects: The Missing Link in Design Systems
soysaucechin
1
1.2k
jQuery: Nuts, Bolts and Bling
dougneiner
66
8.6k
How to Get Subject Matter Experts Bought In and Actively Contributing to SEO & PR Initiatives.
livdayseo
0
210
Documentation Writing (for coders)
carmenintech
77
5.6k
Bridging the Design Gap: How Collaborative Modelling removes blockers to flow between stakeholders and teams @FastFlow conf
baasie
0
730
Reality Check: Gamification 10 Years Later
codingconduct
0
2.3k
Reflections from 52 weeks, 52 projects
jeffersonlam
356
22k
How to Build an AI Search Optimization Roadmap - Criteria and Steps to Take #SEOIRL
aleyda
1
2.2k
Dominate Local Search Results - an insider guide to GBP, reviews, and Local SEO
greggifford
PRO
0
360
Transcript
ݚڀࣨઆ໌ձ 201910݄9 ஜେֶେֶӃɹਤॻؗใϝσΟΞݚڀՊ ਓͱԻͷใֶݚڀࣨɹࢁຊɹ༤
͜ΜΜ
B4ͷΛ ฉ͖͔ͨͬͨਓ
͝ΊΜͳ͍͞
ϐϯνώολʔͰ ͕͠ͳ͍M1͕͕ ൃද͢Δ͜ͱʹ ͳͬͪΌ͍·ͨ͠
ݚڀࣨότϧ ೩͑ͯ·͔͢ʁʁ
ߴͳใઓ ָ͠ΜͰ·͔͢ʁʁ☠
ݚڀࣨଐ ্࢘ΛબΔ ࠷ޙͷػձͩ͠ɼɹ
ݚڀࣨ Ұͱ͍͏ ܾͯ͘͠ͳ͍࣌ؒΛ ա͢͝ͱ͜Ζɹ
ͥͻޙչͷͳ͍ ݚڀࣨଐΛͯ͠Ͷ
ΞδΣϯμ • ࣗݾհɾݚڀࣨʹདྷ͖͔͚ͨͬ • ଔݚͰͬͯͨݚڀ • ݚڀੜ׆ͬͯͲΜͳ͔Μ͡ʁ • Q&A
ࣗݾհ
ࢁຊ ༤ (Yuya Yamamoto) LSPC ฏլάϧʔϓ M1 ʢԻԻڹֶجૅTAͩͬͨΑʂʣ ɹʢใϝσΟΞֶྨ20173࣍ฤೖʣ ઐɿԻɾԻָใॲཧ
ݚڀςʔϚɿՎএٕೳʹணͨ͠Վͱָۂͷੳ ʢݱࡏɼ࢈૯ݚͱڞಉݚڀதʣ ͖ͳͷɿӡɾञ झຯɿਓੜ কདྷͷເɿՈ͔Βग़ͣʹຊΛಡΜͰҰੜΛա͢͝
ݚڀࣨʹདྷ͖͔͚ͨͬ
ଔݚͰԿΛ͔ͨ͠
ՎͷқͰ ۂΛ୳͢γεςϜ
ՎͷқΛࣗಈਪఆ͢ΔγεςϜ Վ͍͍͢ۂ &#$ " ! #% "
) ( *)2 /0 %#"' &1 4 .,3 $(!+ + - ( ) Վʹқ͕͋Δ қࣗʹ߹͏ՎΛ ୳͢ख͕͔ΓͱͳΔ͔ʁ .*%*σʔλ Վ͍ʹ͘͞ ͷࢉग़ ࣗʹ߹͏ՎΛ ͓͢͢Ίͯ͘͠ΕΔγεςϜ ͪ͜Β Վ͍͍͢ ԻҬΊ ͷͼͷͼͨ͠ ϝϩσΟ Վ͏ͱ͖ͷಛͰ ۂΛ୳͢͜ͱ͕Ͱ͖ΔγεςϜ Ԡ༻ྫ
ݚڀՌ ਓʹΑͬͯײ͡ΔՎ͍ʹ͘͞ҧ͏ ݸਓ͝ͱͰқਪఆ͢Δͱྑ͍ ɾٸʹߴ͘ͳΔՕॴ͕͋Δ ɾԻҬ͕͍͋ͬͯͳ͍ ɾϝϩσΟ͕ෳࡶ ɹͷ؍͕͠͞ͷओཁૉ ࢀՃऀ બ͞Εͨม, ()ճؼ
ย ܾఆ 1 ટํมԽׂ߹(2.262), BPM(0.017), ࠷ߴԻ(0.178), ϐονΤϯτϩϐʔ(1.514) -13.29 0.528 2 Լ߱༂ׂ߹(4.817), LR(0.085), ্༂ׂ߹(9.698), ԻՁฏۉ(-1.846) 2.83 0.765 3 WPCS(0.359), BPM(0.299), Իߴฏۉ(0.447), Լ߱༂ׂ߹(11.28), -30.53 0.753 4 ટํมԽׂ߹(3.193), ԻҬ(-0.202), λΠͷ(-0.061) 7.345 0.337 5 ટํมԽׂ߹(2.735), Լ߱༂ׂ߹(6.249) 3.789 0.456 6 λΠͷ(-0.143), ϐονΤϯτϩϐʔ(2.897) LR(0.098), ࠷ߴԻ(-0.151), ԻՁΤϯτϩϐʔ (0.723) 8.51 0.737 7 LR(0.048), BPM(0.011), Լ߱༂ׂ߹(5.178) 1.967 0.421 8 Իߴฏۉ(0.254), ԻՁΤϯτϩϐʔ(2.198) -13.51 0.485 9 ϐονΤϯτϩϐʔ(1.681), λΠͷ(-0.055) 1.883 0.178 10 ԻՁີ(0.035), ԻՁฏۉ(3.558), ࠷ߴԻ(0.179), ԻՁΤϯτϩϐʔ (1.409) -18.38 0.507 11 ࠷ߴԻ(0.196) -8.201 0.237 12 Լ߱༂ׂ߹(15.66), ԻҬ(-0.436), ϐονΤϯτϩϐʔ(4.516) -1.653 0.512
ৄࡉWEBʢจʣͰʂ →IUUQT XXXTMJTUTVLVCBBDKQMTQD*14+ .64QEG
ݚڀੜ׆ʹ͍ͭͯ
͍ΖΜͳελΠϧͷ ݚڀ͕ࣨ͋Γ·͢ΑͶ
ฐݚ ׂͱ์ΑΓ ίΞλΠϜͳ͍
͖ʹݚڀ͕Ͱ͖Δ )PXFWFS ଵ͚͕ͪʹͳΓɼ ଵ͚Δͱ † ࢮ † ͕๚ΕΔ
ࢁຊͷ߹
3݄ 8݄ 11݄ 12݄ 1݄Լ० ʢʒʣ ݚڀςʔϚ͕ ͳ͔ͳ͔ܾ·Βͳͯ͘ ͠ΜͲ͔ͬͨ࣌ظ Γ߹͍Λ͋ͭΊͯ ࣮ݧ͕͏·͍͔͘͘ গਓͰ͓ࢼ͠ ਓΛΑΓ ͨ͘͞Μ ूΊ࣮ͯݧ ੳΛճͭͭ͠ ଔͷࣥච ɿଵ͚ͯࢮΜͩظؒ ɿӃࢼ (2ճ͋Δͷͯ͠)
ςʔϚͷܾ·Γํ
ฐݚͷςʔϚܾఆ γʔζࢤ͕ଟ͍
γʔζʁʁ
ݚڀςʔϚΛܾΊΔํ๏ γʔζ (Seeds) - ࣗͰΓ͍ͨςʔϚΛݟ͚ͭͯ͘Δ χʔζ (Needs) - ςʔϚ্͕͔Β͖߱ͬͯͯͦΕʹऔΓΉ
γʔζͱχʔζͲ͕͍͍ͬͪʁ • χʔζʢ߱ͬͯ͘Δʣ • ͕໌֬ɼڭһͷॿ͚ಘ͍݁͢͠Ռ ग़͍͢ • Β͞ΕͯΔײΛײ͡Δ͜ͱ͋Δ
γʔζͱχʔζͲ͕͍͍ͬͪʁ • γʔζʢࣗͰݟ͚ͭΔʣ • ͭΑ͘ͳΕΔɼࣗͷΓ͍ͨ͜ͱ͔ͩΒ ָ͍͠ • ઃఆ͕݁͘͠ՌΛग़͢ͷ͕ࠔɼɹ ߹ʹΑͬͯڭһͷॿ͚͕ेʹͰ͖ͳ͍ ͜ͱ
ڵຯͷ͋Δ Իؔ࿈ͷςʔϚ͕ ͋Δ ݚڀςʔϚͱͯ͠ ཱͦ͠͏ ͱΓ͋͑ͣԿ͔ ԻͰͬͯΈ͍ͨ ͱΓ͋͑ͣ ฐݚʹߦ͖͍ͨ αʔϕΠ͠Α͏
ςʔϚΛઌੜ͔ΒఏҊ ͞ΕͯͦΕΛΔ ͜ͷݚڀςʔϚͰ (0 ͍ ͍ ͍ ͍ ͍͍͑ ͍͍͑ ͍͍͑ ݟ͔ͭͬͨ ݟ͔ͭΒͳ͍ ͍͍͑ γʔζ χʔζ
݁ߏ γʔζͰམͪண͘ ਓ͕ଟ͍ ʢ7ׂ͘Β͍ʣʢγʔζʣ
ݚڀςʔϚ
Իʹؔ͢Δ͜ͱͳΒ େମͳΜͰͰ͖Δ
Իͷ͜ͱΛΓ͍ͨਓ ฐݚ͕͓͢͢Ίʂ Վͱ͔Իͱ͔ڵຯ͋Δਓͩͱࢁຊ͕ت͢ΔΑʂʂʂʂʂʂ ΊͬͪΌWelcomeͩΑʂʂʂʂʂʂʂʂʂ ϚδͰʂʂʂʂʂʂʂʂʂʂʂʂʂʂ