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
Kaggle Google Quest Q&A Labeling - 23th place s...
Search
Shuhei Goda
February 28, 2020
Technology
4
4.1k
Kaggle Google Quest Q&A Labeling - 23th place solution
Shuhei Goda
February 28, 2020
Tweet
Share
More Decks by Shuhei Goda
See All by Shuhei Goda
Turing × atmaCup #18 - 1st Place Solution
hakubishin3
0
830
ジョブマッチングサービスにおける相互推薦システムの応用事例と課題
hakubishin3
3
920
とある事業会社にとっての Kaggler の魅力
hakubishin3
8
2.4k
課題の解像度が荒かったことで意図した改善ができなかった話
hakubishin3
3
990
Wantedly におけるマッチング体験を最大化させるための推薦システム
hakubishin3
4
1.2k
Recommendation Industry Talks #1 Opening
hakubishin3
1
380
会社訪問アプリ「Wantedly Visit」での シゴトに関する興味選択機能と推薦改善
hakubishin3
0
620
論文紹介: Improving Implicit Feedback-Based Recommendation through Multi-Behavior Alignment(Xin Xin et al., 2023)
hakubishin3
0
610
Feedback Prize - English Language Learning における擬似ラベルの品質向上の取り組み
hakubishin3
0
980
Other Decks in Technology
See All in Technology
MCP Documentation Server @AI Coding Meetup #1
yyoshiki41
2
2.6k
Android는 어떻게 화면을 그릴까?
davidkwon7
0
100
LLM とプロンプトエンジニアリング/チューターをビルドする / LLM, Prompt Engineering and Building Tutors
ks91
PRO
1
210
IVRyにおけるNLP活用と NLP2025の関連論文紹介
keisukeosone
0
180
フロントエンドも盛り上げたい!フロントエンドCBとAmplifyの軌跡
mkdev10
2
250
Devinで模索する AIファースト開発〜ゼロベースから始めるDevOpsの進化〜
potix2
PRO
6
2.8k
似たような課題が何度も蘇ってくるゾンビふりかえりを撲滅するため、ふりかえりのテーマをフォーカスしてもらった話 / focusing on the theme
naitosatoshi
0
390
アジャイル脅威モデリング#1(脅威モデリングナイト#8)
masakane55
3
160
20250413_湘南kaggler会_音声認識で使うのってメルス・・・なんだっけ?
sugupoko
1
400
GitHub MCP Serverを使って Pull Requestを作る、レビューする
hiyokose
2
710
20250411_HCCJP_AdaptiveCloudUpdates.pdf
sdosamut
1
100
ElixirがHW化され、最新CPU/GPU/NWを過去のものとする数万倍、高速+超省電力化されたWeb/動画配信/AIが動く日
piacerex
0
110
Featured
See All Featured
Being A Developer After 40
akosma
91
590k
Gamification - CAS2011
davidbonilla
81
5.2k
RailsConf 2023
tenderlove
30
1.1k
Chrome DevTools: State of the Union 2024 - Debugging React & Beyond
addyosmani
5
520
Reflections from 52 weeks, 52 projects
jeffersonlam
349
20k
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
mongodb
45
9.5k
Scaling GitHub
holman
459
140k
Rails Girls Zürich Keynote
gr2m
94
13k
It's Worth the Effort
3n
184
28k
Fantastic passwords and where to find them - at NoRuKo
philnash
51
3.1k
Building Flexible Design Systems
yeseniaperezcruz
329
38k
Refactoring Trust on Your Teams (GOTO; Chicago 2020)
rmw
34
2.9k
Transcript
©2020 Wantedly, Inc. 23th place solution Kaggle Google Quest Q&A
Labeling লձ Feb 28, 2020 - Shuhei Goda - @jy_msc
©2020 Wantedly, Inc. Team - The Hand Shuhei Goda @jy_msc
Visit Engineering Team at Wantedly Naomichi Agata @agatan_ People Engineering Team at Wantedly
©2020 Wantedly, Inc. Model Pipeline #FSUCBTF VODBTFE -JHIU(#. #FSUCBTF VODBTFE
Settings ɾ3fold with GroupKFold ɾBCE + margin ranking loss ɾ3epoch Settings ɾmax_depth=1 ɾlr=0.1 Meta features ɾtext length ɾstackexchange Text data ɾquestion_title ɾquestion_body ɾanswer 1SF1SPDFTT 2BOE" 1SF1SPDFTT POMZ2 ɾquestion_title ɾquestion_body ɾquestion_title ɾquestion_body ɾanswer Settings ɾhtml escape ɾhead+tail truncation
©2020 Wantedly, Inc. ɾHTMLจࣈྻͷΞϯΤεέʔϓ Pre-Process IUUQTXXXLBHHMFDPNDHPPHMFRVFTUDIBMMFOHFEJTDVTTJPO
©2020 Wantedly, Inc. ɾςΩετσʔλͷ݁߹ͱτϦϛϯά ɹɾ[CLS] + question_title + [SEP] +
question_body + [SEP] + answer ɾquestion_body ͱ answer ͕ࢦఆͷ͞Λ͑ͨ߹, ͔྆ΒಉαΠζΛτϦϛϯά Pre-Process IUUQTBSYJWPSHBCT
©2020 Wantedly, Inc. ɾBert-base (uncased) ɹɾޙΖ4ͭͷӅΕͷग़ྗΛ༻ https://arxiv.org/abs/1905.05583 ɹɾQAؒͷSEP tokenͷग़ྗΛ༻ Model
Architecture
©2020 Wantedly, Inc. ɾLabel weight ɹɾ؆୯ͦ͏ͳλεΫweightΛখ͘͞, ෆۉߧͰͦ͠͏ͳλεΫweightΛେ͖͘ ɹɾgpyoptͰweightͷ୳ࡧΛࢼͨ͠Έ͕ͨ, Լهͷ୯७ͳΓํ͕࠷ྑ͔ͬͨ Loss
function Label weight ͋Γ Public: 0.45979, Private: 0.41440 Label weight ͳ͠ Public: 0.43455, Private: 0.40602
©2020 Wantedly, Inc. ɾBCE + margin ranking loss (1 :
1) ɹɾϛχόονΛ2ͭʹׂͯ͠ margin ranking loss Λܭࢉ Loss function BCE + margin ranking loss Public: 0.45979, Private: 0.41440 BCE Public: 0.44006, Private: 0.40668
©2020 Wantedly, Inc. ɾQuestion Model ɹɾQ༻ͷλεΫΛQuestion text͚ͩΛͬͯղ͘ ɹɾΠϯϓοτQ͚ͩͰ͍͍ͷͰ, Qͷtruncationͷྔ͕ݮΔ (Qͷใྔ͕૿͑Δ)
Training Q model + Q and A model Public: 0.45979, Private: 0.41440 Q and A model × 2 (seed average) Public: 0.44298, Private: 0.40613
©2020 Wantedly, Inc. ɾLightGBM ɹɾmax_depth=1, lr=0.1 ɹɾmeta features ɹɹɾtext length
(question, answer) ɹɹɾmeta data from stackexchange (Score, View, FavoriteCount, …) Post-Process LightGBM Public: 0.45979, Private: 0.41440 Simple binning without meta features Public: 0.45282, Private: 0.41387
©2020 Wantedly, Inc. Why we used LightGBM? 1. Simple binning
method ɹɾ༧ଌΛࢄԽ͢Δ͜ͱͰ Spearman’s correlation ͕ྑ͘ͳΔ͜ͱʹؾͮ͘ ɹɾtarget͝ͱʹϏϯαΠζΛࣄલʹઃఆͯ͠Ϗϯೋϯά ɹɾϏϯαΠζݻఆʹ্ͨ͠ͰBertͷ֤epochͷग़ྗΛweighted average (weight࠷దԽ)
©2020 Wantedly, Inc. Why we used LightGBM? 2. Optimize bin-size
and weights ɹɾϏϯαΠζ࠷దͳΛ͍ͨ͘ͳͬͨ ɹɾϏϯαΠζͱweightsͷಉ࣌࠷దԽ্͕ͨ͠ख͍͔͘ͳ͍ ɹɾ࠷దͳϏϯαΠζ༧ଌͷܗʹΑܾͬͯ·Δ. ֤foldͷ࠷దͳϏϯαΠζͷฏۉͱ weighted averageޙͷ༧ଌ࠷దͳͷ͔Βဃ͢Δ
©2020 Wantedly, Inc. Why we used LightGBM? 3. LightGBM ɹɾϏϯαΠζͱweightsͷಉ࣌࠷దԽ͍ͨ͠
ɹɾmeta features͍͍ͨ ɹɾGBDTσʔλΛׂׂͯ͠ޙͷྖҬʹ࠷దͳΛׂΓͯΔख๏ ɹɹˠ ઙ͍߹Ϗϯχϯάͱಉ༷ͷࢄԽ͕Ͱ͖ΔΜ͡Όͳ͍͔ max_depth=2 max_depth=8
©2020 Wantedly, Inc. 4. LightGBM (parameter tuning) ɹɾࢄԽ͢Δ΄Ͳscore͕ྑ͘ͳΔͷͰ, ߏΛۃྗγϯϓϧʹ͍ͨ͠ ɹɾtrainσʔλΛׂͯ͠࠷దͳύϥϝʔλΛݟ͚ͭΔ
ɹɾmax_depthΛҰ൪খ͘͞, lrΛۃྗେ͖ͨ͘͠ํ͕score͕ྑ͘ͳͬͨ Why we used LightGBM?
©2020 Wantedly, Inc. ɾsample weightͷઃఆ ɾhostͷ୯ޠΛΠϯϓοτͷઌ಄ྻʹஔ͘ ɾnew tokenͷՃ ɾBert-base casedΛ͏
ɾtexͷίʔυϒϩοΫΛྗٕͰফڈ Didn’t work for us
©2020 Wantedly, Inc. Discussion: https://www.kaggle.com/c/google-quest-challenge/discussion/129904#742302 Kernel: https://www.kaggle.com/shuheigoda/23th-place-solusion Links
©2020 Wantedly, Inc. https://www.wantedly.com/projects/375150 We are hiring !