Nowadays, the renaissance of artificial intelligence (AI) has attracted huge attention from every corner of the world. On the one hand, neural algorithms and theories (include shallow and deep ones) have nearly dominated AI development in almost all areas, e.g., natural language processing (NLP), computer vision (CV) and planning and have shown great promise. On the other hand, recommender systems (RS), as one of the most popular and important applications of AI, has been widely planted into our daily life and has made a huge difference. Naturally, the combination of neural algorithms and theories and recommender systems has been flourishing for years and has shown great potential.

In practice, neural models and algorithms have nearly dominated the recommender system research in recent years. Many state-of-the-art recommender systems are built on neural algorithms, especially deep neural algorithms. However, most of existing researchers often only focus on the application of deep neural models to solve the problems in recommender systems, they either ignore the more efficient shallow and light weighted neural models or overlook the fundamental theories behind these neural models, and the intrinsic connections between these theories and the recommender system issues.

This workshop aims to systematically discuss the recent advancements of both shallow and deep neural algorithms for recommender systems from both the application and theoretical perspectives. Particularly, the recent progress achieved in both shallow and deep neural recommender system algorithms together with the related theories will be discussed. Furthermore, both the recent progress achieved in the academia and the industry will be covered. This workshop solicits the latest and significant contributions on developing and applying neural algorithms and theories for building intelligent recommender systems.

The workshop invites submissions on all topics of neural algorithms and theories for recommender systems, including but not limited to:

Deep neural model for recommender systems
Shallow neural model for recommender systems
Neural theories particularly for recommender systems
Theoretical analysis of neural models for recommender systems
Theoretical analysis for recommender systems
Data characteristics and complexity analysis in recommender systems
Non-IID (non-independent and non-identical distribution） theories and practices for recommender systems
Auto ML for recommender systems
Privacy issues in recommender systems
Recommendations on small data sets
Complex behavior modeling and analysis for recommender systems
Psychology-driven user modeling for recommender systems
Brain-inspired neural models for recommender systems
Explainable recommender systems
Adversarial recommender systems
Multimodal recommender systems
Rich context recommender systems
Heterogeneous relations modeling in recommender systems
Various recommendation scenarios including but not limited to collaborative filtering, sequential recommendations, social recommendations, conversational recommendations, news recommendations, music recommendations, etc.
The application of recommender systems in emerging domains including health care, Fintech, education, fashion industry, etc.
Visualization in recommender systems
New evaluation metrics and methods for recommender systems
Interpretable recommendation
Trustworthy recommendation