The Recsys Challenge is one of the most important global competitions in the field of recommender systems and enjoys significant prestige, especially within academic circles and industries related to data analysis and artificial intelligence.
Thank you! Your submission has been received!
By clicking “Stay Updated” button, you agreed to allow Synerise to send information regarding RecSys Challenge 2025 to your email address and thus to process your personal data by Synerise for this purpose.Synerise is the controller of your personal data processed for the purpose of sending updates regarding RecSys Challenge 2025.
You may withdraw your consent at any time by contacting us. For more information on how we process your personal data and what rights you have, see our Privacy Policy.
Oops! Something went wrong while submitting the form.
Past RecSys Challenge co-hosts
Industry leaders with exceptional data science teams
RecSys Data Challenge
Why Participate in the RecSys Challenge?
The ACM RecSys Challenge is an international competition organised by the ACM RecSys Conference in cooperation with a sponsoring company. Over 100 teams from universities and large companies from all over the world challenge each other to find solutions for real problems related to recommender systems (content filtering software that create customised user-specific recommendations to help users in their choices). During the competition, participants apply various data science techniques, such as machine learning, data mining, game theory and soft computing. The best teams are invited to present their work at the conference.
ACM RecSys
The competition is organized as part of ACM RecSys, the most prestigious conference on recommender systems. Publications from this conference are considered milestones in the field. The competition results are often presented at the conference, providing participants with an opportunity to gain recognition in both academic and industry communities.
Collaboration with Industry
The competition’s partners are often well-known tech companies (e.g., Amazon, Spotify, Zalando, Twitter). This means: The tasks reflect real-world challenges in the industry. The solutions from the competition can directly improve recommender systems in actual applications.
Advanced Datasets and Problems
The datasets used in the competition are complex, diverse, and often unique—participants have the opportunity to work with data that is not easily available to the public. The challenges are at the cutting edge of innovation (e.g., sequential recommendations, contextual recommendations, group recommendations, hybrid systems).
Prestige and Career Development
Success in this competition can open many doors, both in academia and industry. The work of top-ranking teams is often published and cited, which is crucial for researchers and academics. In the professional environment, participation (especially achieving top rankings) can be a great addition to a portfolio.
Promoting Innovation
The challenge pushes participants to develop new algorithms and approaches. Some techniques developed during the competition later become standards in the field. For example: Advancements in methods such as Deep Learning for recommendations, sequence-based models (e.g., RNNs), or the inclusion of contextual data.
Networking Opportunities
The competition attracts both beginners and experts in the field. For participants, it’s an excellent opportunity to network with leading researchers, companies, and institutions.
RecSys Challenge 2025
The challenge is designed to promote a unified approach to behavior modeling. Many modern enterprises rely on machine learning and predictive analytics for improved business decisions. Common predictive tasks in such organizations include recommendation, propensity prediction, churn prediction, user lifetime value prediction, and many others. A central piece of information that is used for these predictive tasks are logs of past behavior of users e.g., what they bought, what they added to their shopping cart, which pages they visited. Rather than treating these tasks as separate problems, we propose a unified modeling approach.
To achieve this, we introduce the concept of Universal Behavioral Profiles—user representations that encode essential aspects of each individual’s past interactions. These profiles are designed to be universally applicable across multiple predictive tasks, such as churn prediction and product recommendations. By developing representations that capture fundamental patterns in user behavior, we enable models to generalize effectively across different applications.
2025
Challenge Task
The objective of this challenge is to develop Universal Behavioral Profiles based on the provided data, which includes various types of events such as product buy, add to cart, remove from cart, page visit, and search query. These user representations will be evaluated based on their ability to generalize across a range of predictive tasks. The task of the challenge participants is to submit user representations, which will serve as inputs to a simple neural network architecture. Based on the submitted representations, models will be trained on several tasks, including some that are disclosed to participants, called "open tasks," as well as additional hidden tasks. The final performance score will aggregate results from all tasks. We iterate model training and evaluation automatically upon submission, the task of the participants is to submit universal user representations.
Open Tasks:
Churn Prediction
Using the submitted user representations, the model will predict whether an active user (defined as one with at least one product-buy event) will churn, i.e., make no purchases within the next 14 days.
Product Propensity
Using the submitted user representations, the model will predict products which a user is most likely to purchase within the next 14 days, from a predefined subset of items.
Category Propensity
Using the submitted user representations, the model will predict the categories in which a user is most likely to make a purchase within the next 14 days, from a predefined subset of categories.
Hidden Tasks:
In addition to the open tasks, the challenge includes hidden tasks, which remain undisclosed during the competition. The purpose of these tasks is to ensure that submitted Universal Behavioral Profiles are capable of generalization rather than being fine-tuned for specific known objectives. Similar to the open tasks, the hidden tasks focus on predicting user behavior based on the submitted representations, but they introduce new contexts that participants are not explicitly optimizing for. After the competition concludes, the hidden tasks will be disclosed along with the corresponding code, allowing participants to replicate results.
The NVIDIA Merlin and KGMON team earned 1st place in the RecSys Challenge 2021 by effectively predicting the probability of user engagement within a dynamic environment and providing fair recommendations on a multi-million point dataset.
The Intel RecSys2023 team ranked #2 in the industry track of RecSys Challenge 2023, presenting a state-of-the-art privacy-preserving recommendation system with graph-enhanced feature engineering.
DeNA data scientist team wins RecSys Challenge 2024
A team of data scientists from DeNA won the RecSys Challenge 2024, a competition held in conjunction with the internationally prestigious 18th ACM Conference on Recommender Systems (RecSys).
— The world’s most advanced behavioral AI infrastructure
Synerise is a deep-tech company specializing in artificial intelligence and big data, founded in 2013 in Kraków, with additional offices in Warsaw, San Francisco, and Dubai. The company provides advanced, proprietary AI and big data solutions across over 150 markets, serving industry leaders in retail, banking, e-commerce, automotive, insurance, and telecommunications. Synerise is a multi-award-winning company, recognized for its innovation and groundbreaking contributions to AI and big data. The company has received numerous international accolades for its technological achievements and has been repeatedly honored for its advanced behavioral modeling solutions, which push the boundaries of what AI can achieve. Synerise processes transactions worth over €150 billion annually and handles trillions of operations each year. Its platform executes more than 12 billion automated decisions monthly, of which 4 billion are supported by artificial intelligence. Key technological advantages of Synerise include: A. Terrarium: one of the fastest proprietary real-time database engines. B. basemodel.ai: the world's most powerful private foundational model for behavioral data. C. Synerise Self-Service Experience: a platform offering hundreds of ready-to-deploy business scenarios from day one. Being selected as the content organizer of the RecSys Challenge 2025 represents an extraordinary opportunity for Synerise to contribute to the global advancement of AI technologies and showcase the potential of Polish innovation on the international stage.
.png)
