Encode Me If You Can: Learning Universal User Representations via Event Sequence Autoencoding

2025-08-11

AI Summary

To address the inefficiency of repetitive feature engineering and model training in multi-task user modeling, this paper proposes a generic user behavior representation framework. It uniformly encodes heterogeneous user interactions (e.g., clicks, purchases) into temporal event sequences and employs a GRU-based autoencoder to learn fixed-dimensional dense vectors, enabling effective sequence compression and reconstruction. Furthermore, it integrates multi-granularity embeddings—including positional, interaction-type, and temporal embeddings—to construct a robust, unified user representation. Crucially, the framework requires no task-specific adaptation and can be directly transferred to downstream tasks such as churn prediction, recommendation, and lifetime value (LTV) estimation. Evaluated on the RecSys Challenge 2025, it achieved second place, demonstrating strong cross-task generalization, computational efficiency, and scalability.

Technology Category

• Machine Learning: Representation Learning
• Data Mining & Knowledge Management: Recommender Systems
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Building universal user representations that capture the essential aspects of user behavior is a crucial task for modern machine learning systems. In real-world applications, a user's historical interactions often serve as the foundation for solving a wide range of predictive tasks, such as churn prediction, recommendations, or lifetime value estimation. Using a task-independent user representation that is effective across all such tasks can reduce the need for task-specific feature engineering and model retraining, leading to more scalable and efficient machine learning pipelines. The goal of the RecSys Challenge 2025 by Synerise was to develop such Universal Behavioral Profiles from logs of past user behavior, which included various types of events such as product purchases, page views, and search queries. We propose a method that transforms the entire user interaction history into a single chronological sequence and trains a GRU-based autoencoder to reconstruct this sequence from a fixed-size vector. If the model can accurately reconstruct the sequence, the latent vector is expected to capture the key behavioral patterns. In addition to this core model, we explored several alternative methods for generating user embeddings and combined them by concatenating their output vectors into a unified representation. This ensemble strategy further improved generalization across diverse downstream tasks and helped our team, ai_lab_recsys, achieve second place in the RecSys Challenge 2025.

Problem

• Develop universal user representations from behavior logs
• Create task-independent embeddings for diverse predictive tasks
• Reconstruct user event sequences using autoencoder models

Innovation

• GRU-based autoencoder for sequence reconstruction
• Fixed-size vector captures key behavioral patterns
• Ensemble strategy combines multiple embedding methods