I am a Data Scientist and Data Scientist Team Manager at CyberAgent, Inc.

I am interested in Econometrics, Causal Inference, A/B test and Reinforcement Learning. Kaggle Master

• Data Science
• data science team management / building
• econometrics
• causal inference
• A/B test
• Reinforcement Learning
• Business Applications of LLMs
• Python, Scala
• Kaggle

• AI Predictor, 2018.06 - 2018.10
  • Building prediction models for ad platform
• CA Dyve, 2018.10 - 2019.09
  • Building prediction models for ad platform
  • A/B testing
  • Ad Management
• Dynalyst, 2019.09 - 2023.10
  • Data Science Team Manager
  • A/B testing
  • Building an A/B testing platform for ad creatives, Bandit Algorithms
  • Building recommendation engines, prediction models
• GAIN Ads 2021.12 - 2023.10
  • Project Manager / Data Science Team Manager
  • New Business Strategy: Partnered with the General Manager to lead the strategic roadmap and development of "GAIN ADS" (a new User Acquisition DSP).
  • Cross-functional Leadership: Orchestrated collaboration between Business, Engineering, and Data Science units to execute a high-performance bidding strategy.
  • Revenue Impact: Successfully scaled the initiative into a core revenue pillar for the division, expanding the product portfolio beyond retargeting.
• Prism Partner, 2023.4 - present
  • data science team manager
  • Development, Feature Implementation, and System Design of Targeted Advertising Delivery Systems
  • Development, Feature Implementation, and System Design of Retail Promotional Advertising Delivery Systems
  • Implementation of Advanced Technology in Society (Specific Achievement: Accepted for WSDM Industry Day)
  • Operations Cost Reduction and New Feature Planning through LLM Applications
• Data Science Center, 2021.10 - present
  • Team Leader for Internal Evaluation and Professional Development Initiatives

• Department of Statistics, Graduate school of Economics, University of Tokyo(M.A.), 2016.04 - 2018.03
  • Research Topics: Causal Inference(RDD), Econometrics, Statistics
  • Supervisor: Prof. Katsumi Shimotsu
• Department of Economics, Graduate school of Economics, University of Tokyo(B.A.) 2012.04 - 2016.03
  • Research Topics: Econometrics, Game Theory
  • Supervisor: Prof. Katsumi Shimotsu

• 伊藤寛武, 金子雄祐, 安井翔太: 『Pythonで学ぶ効果検証入門』 2024 (Amazon Link)

• Daiki Katsuragawa, Yusuke Kaneko, Kaito Ariu, Kenshi Abe. Efficient Creative Selection in Online Advertising using Top-Two Thompson Sampling. WSDM (Industry Day). 2025
• Abe, Kenshi, and Yusuke Kaneko. "Off-Policy Exploitability-Evaluation in Two-Player Zero-Sum Markov Games." AAMAS 2021 (Full Paper)
• Morishita, Gota, et al. "Online Learning for Bidding Agent in First Price Auction." Workshop on Reinforcement Learning in Games in Thirty-Fourth AAAI Conference on Artificial Intelligence. 2020.

• Kato, Masahiro, and Yusuke Kaneko. "Off-policy evaluation of bandit algorithm from dependent samples under batch update policy." arXiv preprint arXiv:2010.13554 (2020).

• 伊藤寛武, 金子雄祐: "リターゲティング広告配信における不連続回帰を用いたリフト効果分析." 人工知能学会全国大会論文集 第 36 回 (2022). 一般社団法人 人工知能学会, 2022.(全国大会優秀賞受賞)
• 阿部拳之, 金子雄祐: “二人零和マルコフゲームにおけるオフ方策評価のためのQ学習”, 第25回ゲームプログラミングワークショップ, 2020.

https://www.kaggle.com/ykaneko1992

Kaggle Master

• Santander Value Prediction Challenge(link) 12/4463
• PLAsTiCC Astronomical Classification(link) 27/1089
• PetFinder.my Adoption Prediction(link) 36/2023
• Predicting Molecular Properties(link) 90/2737
• Home Credit Default Risk(link) 145/7176
• AI Mathematical Olympiad - Progress Prize 1(link) 64/1161

• Pythonで学ぶ効果検証入門 ー Forkwell Library #57
• 因果推論の計量経済学刊行記念トークイベント(link)

• "clustering for private interest-based advertising" & "learning a logistic model from aggregated data"
  • KDD2021 参加報告&論文読み会 2021/09/24 (slide)
• 広告クリエイティブ最適化と評価のためのBandit
  • CF + FinML勉強会 2021/03/13 (slide)

• ビジネス(の人)的に嬉しいコンペ開催のやり方
  • Discovery DataScience Meet up (DsDS) #1 2020/11/13 (slide)

• 企業の中の経済学
  • 東京大学駒場祭 (slide)

• 研究と事業の融合がもたらす革新― バンディットアルゴリズムの社会実装を成功に導いた舞台裏
• 世界が認めた技術で革新する広告ビジネス　Prism Partnerの挑戦
• AI事業本部のデータサイエンティスト伊藤 寛武・金子 雄祐が執筆し、AI Lab研究員の安井 翔太が監修した「Pythonで学ぶ効果検証入門」が発売開始
• AI事業本部「Dynalyst」のデータサイエンティスト伊藤 寛武・金子 雄祐による論文が人工知能学会全国大会(JSAI2022)において全国大会優秀賞を受賞
• AI Lab、マルチエージェント分野のトップカンファレンス「AAMAS 2021」にて論文採択
• ビジネスの現場で価値を出すデータサイエンティストへーサイバーエージェントの「DSOps研修」とはー
• 新卒データサイエンティスト研修「DSOps研修」とは何か?
• CA × atmaCup 2ndを開催しました
• CA × atmaCupを開催しました & 開催準備悲喜こもごも
• アドテクスタジオのインターン「アドテクコンペ」を開催しました