){kind=avatar}
Shunsuke Kitada
Research Scientist working on Vision & Language with Deep Learning
Research Scientist @Computer Vision Lab, LINE Corp.
23th Floor Yotsuya Tower, 1-6-1 Yotsuya, Shinjuku-ku, Tokyo 160-0004, Japan
Project Researcher @Hosei University
S603, 3-7-2, Kajino-cho, Koganei-shi, Tokyo 184-8584, Japan
Intelligent information processing laboratory
Biography
Shunsuke Kitada （北田 俊輔 in Japanese) is a Research Scientist at Computer Vision Lab (CVL), LINE Corp. His research interest is now on computational advertising with a focus on automatic geneartion/evaluation and assistive technology for multi-modal ad creatives.
Previously, he got his Ph.D. in 2023 at major in applied informatics, graduate school of science and engineering, Hosei University under the supervision of Prof. Hitoshi Iyatomi. His project was to improve prediction performance and model interpretability through attention mechanisms from basic and applied research perspectives.
His personality traits can be summarized as follows:
- ❤️ Love research and development. He is enjoying research life, and he is currently focusing on three research topics simultaneously, e.g., natural language processing, medical image-based computer vision, computational advertising.
- 📝 Every day read and implement the SoTA models. He has released many re-implementations of models using mainly Chainer and PyTorch. Therefore, based on state-of-the-art cases, he can advise on deep learning-based product design.
- 😄 High technical communicativity. Summarize what he made and what he studied, and spread such information. This shows that he can input and output regularly.
The resume is available in PDF .
Interests
- 🤖 Natural Language Processing
- 💻 Computer Vision
- 🏥 Medical Image Processing
- 📃 Computational Advertising
Education
-
PhD in Engineering, 2023
Graduate School of Science and Engineering, Hosei University
-
MSc in Engineering, 2020
Graduate School of Science and Engineering, Hosei University
-
BSc in Engineering, 2018
Hosei University
Recent News 🗞️
Conference Paper (Refereed) 📝
Domestic Conference / Presentation in Japan 🇯🇵
(2023).
Majority or Minority: 固有表現抽出におけるデータの不均衡性に着目した損失関数の提案.
NLP 2023.
(2022).
固有表現認識タスクにおけるデータセットの偏りに着目した動的重み付け損失関数の提案.
YANS 2022.
(2022).
Document AI タスクに向けた大規模事前学習済みモデルを活用した Layout-aware Prompting.
YANS 2022.
(2022).
ViT-CLT: パッチ分割した文字画像から偏旁冠脚を考慮した文書分類.
NLP 2022.
(2022).
書き手と読み手の怒りの感情認識の差の原因となる表現の獲得.
NLP 2022.
(2021).
文章の書き手が持つ感情と読み手が受ける感情の差に関する初期調査および考察.
YANS 2021.
(2021).
効果の高いランディングページ作成支援に向けた広告テキストからの効果予測および推薦モデルの構築.
YANS 2021.
(2021).
広告クリエイティブ評価のための深層確率埋め込みの学習.
JSAI 2021.
(2021).
キャプション生成時低品質データ事前検知の試み.
NLP 2021.
(2021).
半教師あり文書分類のための仮想敵対的学習による注意機構の頑健性および解釈性の向上.
NLP 2021.
Recent & Upcoming Talks 🎙️
Awards & Grants 🏆
Posts & Articles 📝
about awards and interviews (almost in 🇯🇵)
Experience 💻
Research Internship
- Supervisor: Mr. Masayoshi Kondo and Dr. Yoshihisa Ijiri
- Topic: Vision-and-language models for advertising
- Worked on establishing a new technology for automatic generation of advertising videos using images
Research Internship
- Supervisor: Dr. Kyosuke Nishida
- Topic: Vision-and-language document understanding
- Worked on establishing a new pre-training method for document image understanding.
Collaborative Researcher
- Supervisor: Dr. Mayu Otani and Dr. Naoto Inoue
- Working on establishing multi-modal deep learning model for ad creative understanding in a special research position for doctoral students adopted as JSPS Research fellow (DC1, DC2, PD).
Research Internship
- Supervisor: Yuki Iwazaki, Researcher
- Topic: Multi-modal advertising (Ad) and landing page (LP) understanding
- Investigated fundamental research to evaluate and analyze ad creatives and its landing pages, presented at academic symposium in Japan.
- My presentations and publications can be found at the CyberAgent tag.
Research Fellowship for Young Scientists (DC2)
Working on towards robust and interpretable deep learning models and its evaluation.
- Research title: KAKENHI-PROJECT-21J14143
- Development of Perturbation Robust and Interpretable Deep Learning Models and Evaluation of Their Interpretability (ja: 摂動に頑健で解釈可能な深層学習モデルの開発とその解釈性の評価）
- Document review & interview area
- Informatics (ja: 情報学）
- Screening section
- Human informatics and related fields (ja: 人間情報学およびその関連分野）
- Subsection
- Intelligent Informatics and Related Fields (ja: 知能情報学関連）
- Area of specialization
- Natural language processing (ja: 自然言語処理）
Application information is based on the page of OIST Groups. Keywords of each field are based on the table.
AI x Ad Consultant
- Surveyed methods for automatically creating/generating ad creatives
- Developed and evaluated prototypes using the method
- Integrated the method into the system, including testing and operating
- Advised on the improvement of advertising creative creation methods
- Participated in meetings (about 1h) at least once a week
- Wrote an article about AI x Adtech in Japanese:
Research Assistant
Part-time Job
- Supervisors: Prof. Masaru Tanaka, Dr. Noriko Umegaki, Prof. Hitoshi Iyatomi
- Assisted in collecting skin images for the construction of an automated diagnostic system.
Teaching Assistant
Assisted in several classes at Hosei University:
Internship
- Worked on quantifying and visualizing doctors' interest from browsing history.
- Built a system for recommending articles for doctors based on the analysis from scratch
Deep Learning Advisor
- Advised how to solve practical problems of existing services that use machine learning algorithms and deep learning models, showing examples of the kind of problems recent deep learning models are capable of solving.
Research Internship
- Supervisor: Dr. Yoshifumi Seki
- Worked on fundamental research to evaluate and analyze ad creatives, resulting in a presentation at Applied Science track of KDD, a top data mining conference.
- My presentations and publications can be found at the Gunosy tag.
- Wrote some articles in Japanese:
- https://data.gunosy.io/entry/ocr-tesseract-deeplearning
- https://data.gunosy.io/entry/yans2018
- https://data.gunosy.io/entry/research-intern-kdd19
- https://data.gunosy.io/entry/adkdd2019-tencent-ads-invited-talk
- https://data.gunosy.io/entry/kdd2019-fairness-and-explainable-ai
- https://data.gunosy.io/entry/can-you-reproduce-the-experiment-pyenv-poetry
Internship
Implemented machine learning models that accurately capture semantic features of a document in a document similarity (Engineer internship 3days / ¥100,000).
Part-time Job
- Worked on improving the user experience and participated in improving the logic of the article distribution.
- Worked on analyzing log data through visualization of multiple KPIs and contributing to the growth of the service.
Teaching Assistant for Deep Learning Hands-On Training Lab
Assisted in several workshops in Tokyo relating to deep learning and CUDA:
Part-time Job
Worked on dealing with large-scale nonstructural patent information in various forms, e.g., pre-processing, crawling, scraping, and analyzing these data.
Fundamental Information Technology Engineer
Fundamental Information Technology Engineer Examination is a yardstick for measuring IT knowledge and skills as a team member by asking a range of questions about algorithm, network, database, information security, practical programming, etc.
Internship
Planned and implemented enterprise resource planning (ERP) packages.
Projects 📂
*
