smly face

Kohei Ozaki

kaggle twitter linkedin github

I am working for Rist inc as a AI Fellow and Senior Manager, focusing on the creation and application of Vision AI.

Experience

Data Science Competitions

I am a Kaggle Grandmaster (an honor for top competitors with outstanding data science skills) with 20 gold medals🏅 in Kaggle competitions. My highest Kaggle rank is 4th out of 530,000+ data scientists as of April 2016.

Below is the list of my prize-winning contests.

Programming Contests

ACM/ICPC is an annual multi-tiered competitive programming competition among the universities of the world. I joined this contest with my friends at Tokyo University of Science and advanced to the Asia Regional Contest.

ICPC

Projects

Mahjong AI

As a hobby project, I have been developing a Mahjong AI using reinforcement learning since 2022. I have also opened a Mahjong AI competition website, "Riichi Lab".

Riichi Lab allows users to participate in the competition by developing and submitting a bot program that runs on a specific Docker image. The submission programs compete against each other, and their rankings are determined by Elo ratings.

Publications

Selected Papers

To see a full list of publications, see my Google Scholar page.

  • Shuhei Yokoo, Kohei Ozaki, Edgar Simo-Serra, Satoshi Iizuka, "Two-stage discriminative re-ranking for large-scale landmark retrieval", Proceeding on the CVPR'20 workshop on Image Matching: Local Features and Beyond, 2020. [arxiv]
  • Yusuke Niitani, Toru Ogawa, Shuji Suzuki, Takuya Akiba, Tommi Kerola, Kohei Ozaki, Shotaro Sano, "Team PFDet's Methods for Open Images Challenge 2019", 2019. [arXiv:1910.11534]
  • Kohei Ozaki, Shuhei Yokoo, "Large-scale landmark retrieval and recognition under a noisy and diverse dataset ", Technical report for Second Landmark Recognition Workshop to be held at CVPR'19, June 16th, 2019. [arxiv][github][dataset][slide1, slide2]
  • Kohei Ozaki, Masashi Shimbo, Mamoru Komachi, Yuji Matsumoto, "Using the mutual k-nearest neighbor graphs for semi-supervised classification on natural language data", Proceedings of the Fifteenth Conference on Computational Natural Language Learning (CoNLL'21), 2011.

Recent Activities

Talks (internatioal conference)

2023-06-19
"5th place solution: kNN shortlist and rotation correction", CVPR'23 WS (Image Matching: Local Features & Beyond), Vancouver
2018-06-18
"Local vs Global Descriptor? Relevance Scoring with using Both", CVPR'18 WS (Large-Scale Landmark Recognition: A Challenge), Salt Lake City
2015-08-09
"3-Stage Ensemble and Feature Engineering for MOOC Dropout Prediction", KDD'15 WS (KDD CUP 2015 - Predicting dropouts in MOOC), Sydney

Lecture, Tutorial, Domestic conference, Meetup events, etc.

2023-09-23
"IMC 2023 振り返り & コードコンペの戦い方", 関東Kaggler会
2023-03-22
"「Kaggleに挑む深層学習プログラミングの極意」著者によるAMA", Weights & Biases 東京ミートアップ #2
2022-12-08
"機械学習コンテストの登り方", ViEW 2022 ビジョン技術の実利用ワークショップ
2018-07-24
"ビッグデータ解析", ゲスト担当分講義, 大阪大学
2018-05-19
"データ分析コンテストの解答から学ぶ", PyCon mini Osaka
2018-03-03
"データ分析コンテストの 勝者解答から学ぶ", 千葉工業大学 ステアラボ人工知能シンポジウム 2018
2017-10-20
"データ分析コンテストの技術と最近の進展", 千葉工業大学 第14回ステアラボ人工知能セミナー
2017-07-29
"画像処理コンペティションにおける技術とワザ", Kaggle Student Meetup by Sansan
2016-09-09
"Python とデータ分析コンテストの実践", FIT2016 第15回情報科学技術フォーラム (富山大学)
2016-07-04
"実験の再現性と効率化の話（Docker と Serialization 周辺）", ビッグデータ基盤技術勉強会 (筑波大学)
2016-03-05
"Workflow, Serialization & Docker for Kaggle", Kaggle Tokyo Meetup #1
2015-10-27
"Techniques (Tricks) for Data Mining Competition", BrainPad 社内講演
2015-10-23
"データマイニングコンテストにおけるテクニック", PyData.Tokyo Meetup #6
2015-10-15
"競技データマイニングにおけるテクニックと動向", 鹿島研究室公開セミナー at 京都大学
2013-11-30
"InfluxDB & LevelDB Inside-out", Monitoring Casual Talk in Kyoto
2012-06-09
"Distributed Graph System & Related Topics", 第18回 データマイニング+WEB 勉強会

Books and Writings

2023-02-02
"Kaggleに挑む深層学習プログラミングの極意", KS情報科学専門書, 講談社 (amazon)
2016-02-01
"国際チームにおけるモデリング手法とチームガイドライン", 情報処理学会誌 (2016年2月号) 小特集
© 2024 ho.lc