mercari AI


Mercari AI’s research “Layout Optimizer for Personalized Home Screen based on Contextual Multi-Armed Bandit in C2C Marketplace” accepted to RecSys 2023’s Workshop


We are pleased to announce that the paper "Layout Optimizer for Personalized Home Screen based on Contextual Multi-Armed Bandit in C2C Marketplace" has been accepted to the CONSEQUENCES '23 workshop within the international conference "RecSys 2023(ACM Conference on Recommender Systems)".

RecSys is one of the most prestigious international conferences in the field of recommendation systems, held annually by researchers from around the world. This year's RecSys, in its 17th year, was held in Singapore from September 18 to 22, 2023.

Key points of presentation

In this presentation, based on how customers use Mercari, we propose a method to display more attractive content on the home screen with priority among other content, including product recommendations. We reported the results of our experiments.

In the proposed method, an exploratory screen display and machine learning are performed to collect data, without giving prior knowledge of what kind of content is attractive to which users. As a result, the screen display is automatically optimized to improve overall satisfaction.


At Mercari app, several teams, including the Recommendation Team, are developing content on the home screen.There are various types of content on the home screen, including item recommendations and campaign announcements. Furthermore, even if the same item is being recommended, there are multiple methods of recommendation and display.

Not only do the items to be recommended vary from user to user, but also the type of recommendation method and how effective the campaign announcement is depends on each user.

Summary of paper

In e-commerce services, including Mercari, the home screen is an important function that attracts many views. However, the amount of content that can be displayed is limited, and it is common practice to place attractive content at the top of the home screen as a priority.

However, since different users have different perspectives on what kind of content is attractive, it is necessary to create a personalized screen structure.

In this study, we proposed a method that applies the contextual multi-armed bandit problem algorithm and reported the details and results of our experiments.

Comments from author

ML Engineer / Backend Engineer | Yusuke Shido

Since this year's event was held offline, and in the form of a poster presentation, we were able to take the time to have direct discussions with researchers and engineers from various countries. In particular, I received a lot of interest and questions from developers and researchers belonging to companies, and I felt that the content of our paper was an issue common to many different companies. The five days at RecSys, including the workshop, were an opportunity for very meaningful input. We hope to apply the knowledge gained to Mercari's business in the future and continue to disseminate the results.

Members making a poster presentation in Singapore

About recommendation team

Mercari's recommendation team is working to improve the quality of product recommendations on each screen of the Mercari App, as well as the mechanisms involved in screen configurations such as the one in this study.


ML Engineer / Backend Engineer

Yusuke Shido

  • Deep Learning
  • Multimodal
  • NLP

Since joining in 2019 Yusuke has been working on a Machine Learning model and backend development in the Customer Reliability Engineering field, such as violation detection systems. Since then he now handles data analysis and implementation for a team that finds tasks and develops solutions in quick iterations. He has been part of a research and development startup while still in school, and is good at coming up with solutions for difficult problems and tasks.

ML Engineer

Shinya Yaginuma

  • Experimentation
  • Personalization

Shinya joined Mercari in 2020. As a data analyst, he worked on workflow improvement related to A/B testing and analysis to improve personalization features. Currently, he is in charge of personalization feature development as an ML Engineer.

ML Engineer


  • Data Science
  • Multimodal
  • Python

Naoya joined Mercari in 2022 as a ML Engineer. As a member of the Recommendation Team, he strives to improve the user experience on Mercari’s home screen. Before joining Mercari, he worked as a data scientist in charge of developing features using analytics and machine learning in an automotive business. He is a Kaggle Master and occasionally participates in competitions.

Product Manager

Tomohiro Furusawa

  • Library Information Science
  • Personalization
  • Recommendation
  • Search

Tomohiro majored in Library and Information Science in graduate school and has been working on product development for information retrieval and recommendation since graduation.
He joined Mercari in May 2019, worked on prototype development of search evaluation and recommendation systems.Then, he launched the recommendation team and is currently a product manager driving development that improves the overall discovery experience.

ML Engineer

Ryo Tanaka

  • MLOps
  • Recommendation

Ryo joined Mercari in 2020. As a software engineer, she has experience developing content moderation and recommendation systems using machine learning. Currently, he is the Tech Lead of the Recommendation Team, working to improve the user experience through personalization.