MEGG: Replay via Maximally Extreme GGscore in Incremental Learning for Neural Recommendation Models

2025/09/09 by Yunxiao Shi, Shuo Yang, Shi, Yunxiao +11
Computer Science · #Artificial Intelligence (cs.AI) #Domain Adaptation and Few-Shot Learning #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Recommender Systems and Techniques

paper · pdf · doi:10.48550/arxiv.2509.07319

openalex publication_date 2025/09/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Neural Collaborative Filtering models are widely used in recommender systems but are typically trained under static settings, assuming fixed data distributions. This limits their applicability in dynamic environments where user preferences evolve. Incremental learning offers a promising solution, yet conventional methods from computer vision or NLP face challenges in recommendation tasks due to data sparsity and distinct task paradigms. Existing approaches for neural recommenders remain limited and often lack generalizability. To address this, we propose MEGG, Replay Samples with Maximally Extreme GGscore, an experience replay based incremental learning framework. MEGG introduces GGscore, a novel metric that quantifies sample influence, enabling the selective replay of highly influential samples to mitigate catastrophic forgetting. Being model-agnostic, MEGG integrates seamlessly across architectures and frameworks. Experiments on three neural models and four benchmark datasets show superior performance over state-of-the-art baselines, with strong scalability, efficiency, and robustness. Implementation will be released publicly upon acceptance.

Citations

• A Learnable Agent Collaboration Network Framework for Personalized Multimodal AI Search Engine
• LLM-Rec: Personalized Recommendation via Prompting Large Language Models
• Large Language Models are Zero-Shot Rankers for Recommender Systems
• Selective and Collaborative Influence Function for Efficient Recommendation Unlearning
• Data-centric Artificial Intelligence: A Survey
• Dataset Pruning: Reducing Training Data by Examining Generalization Influence
• Interpretable Data-Based Explanations for Fairness Debugging
• Towards Continual Knowledge Learning of Language Models
• Recent Advances of Continual Learning in Computer Vision: An Overview
• Incremental Learning for Personalized Recommender Systems
• Online Continual Learning in Image Classification: An Empirical Survey
• A Comprehensive Study of Class Incremental Learning Algorithms for Visual Tasks
• A Practical Incremental Method to Train Deep CTR Models
• ADER: Adaptively Distilled Exemplar Replay Towards Continual Learning for Session-based Recommendation
• Knowledge Distillation: A Survey
• Data Cleansing for Models Trained with SGD
• Deep Learning Recommendation Model for Personalization and Recommendation Systems
• Three scenarios for continual learning
• End-to-End Incremental Learning
• Revisiting Distillation and Incremental Classifier Learning
• Incremental Learning of Object Detectors without Catastrophic Forgetting
• Deep & Cross Network for Ad Click Predictions
• Measuring Catastrophic Forgetting in Neural Networks
• Neural Factorization Machines for Sparse Predictive Analytics
• Deep Interest Network for Click-Through Rate Prediction
• Continual Learning with Deep Generative Replay
• Neural Collaborative Filtering
• Understanding Black-box Predictions via Influence Functions
• DeepFM: A Factorization-Machine based Neural Network for CTR Prediction
• iCaRL: Incremental Classifier and Representation Learning
• Incremental Factorization Machines for Persistently Cold-starting Online Item Recommendation
• Wide & Deep Learning for Recommender Systems
• SLMRec: Distilling Large Language Models into Small for Sequential Recommendation
• LLM-based Medical Assistant Personalization with Short- and Long-Term Memory Coordination
• Gradient-based learning applied to document recognition

Related

• When Newer is Not Better: Does Deep Learning Really Benefit Recommendation From Implicit Feedback?
• ADER: Adaptively Distilled Exemplar Replay Towards Continual Learning for Session-based Recommendation
• Structure Aware Incremental Learning with Personalized Imitation Weights for Recommender Systems
• Memory Augmented Neural Model for Incremental Session-based Recommendation
• Adaptive Neural Architectures for Recommender Systems