# Hierarchical Temporal Convolutional Networks for Dynamic Recommender Systems

arXiv: https://arxiv.org/abs/1904.04381 · HTML: https://mlsanity.com/paper/1904.04381/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Apr 2019.

## Built on by (most cited first, 28 of 28)

- [Identity-aware Graph Neural Networks](https://mlsanity.com/paper/2101.10320/) (2021)
- [ROLAND: Graph Learning Framework for Dynamic Graphs](https://mlsanity.com/paper/2208.07239/) (2022)
- [LLM4DyG: Can Large Language Models Solve Spatial-Temporal Problems on Dynamic Graphs?](https://mlsanity.com/paper/2310.17110/) (2023)
- [Dynamic Graph Collaborative Filtering](https://mlsanity.com/paper/2101.02844/) (2021)
- [A Comprehensive Survey of Dynamic Graph Neural Networks: Models, Frameworks, Benchmarks, Experiments and Challenges](https://mlsanity.com/paper/2405.00476/) (2024)
- [Environment-Aware Dynamic Graph Learning for Out-of-Distribution Generalization](https://mlsanity.com/paper/2311.11114/) (2023)
- [Heterogeneous Global Graph Neural Networks for Personalized Session-based Recommendation](https://mlsanity.com/paper/2107.03813/) (2021)
- [Spectral Invariant Learning for Dynamic Graphs under Distribution Shifts](https://mlsanity.com/paper/2403.05026/) (2024)
- [Hierarchical State Space Models for Continuous Sequence-to-Sequence Modeling](https://mlsanity.com/paper/2402.10211/) (2024)
- [Attention Mixtures for Time-Aware Sequential Recommendation](https://mlsanity.com/paper/2304.08158/) (2023)
- [M2TRec: Metadata-aware Multi-task Transformer for Large-scale and Cold-start free Session-based Recommendations](https://mlsanity.com/paper/2209.11824/) (2022)
- [Diverse Preference Augmentation with Multiple Domains for Cold-start Recommendations](https://mlsanity.com/paper/2204.00327/) (2022)
- [Implicit Session Contexts for Next-Item Recommendations](https://mlsanity.com/paper/2208.09076/) (2022)
- [Transformers Meet ACT-R: Repeat-Aware and Sequential Listening Session Recommendation](https://mlsanity.com/paper/2408.16578/) (2024)
- [IDNP: Interest Dynamics Modeling using Generative Neural Processes for Sequential Recommendation](https://mlsanity.com/paper/2208.04600/) (2022)
- [DTFormer: A Transformer-Based Method for Discrete-Time Dynamic Graph Representation Learning](https://mlsanity.com/paper/2407.18523/) (2024)
- [Everyone's Preference Changes Differently: Weighted Multi-Interest Retrieval Model](https://mlsanity.com/paper/2207.06652/) (2022)
- [HiPAL: A Deep Framework for Physician Burnout Prediction Using Activity Logs in Electronic Health Records](https://mlsanity.com/paper/2205.11680/) (2022)
- [Hierarchical Multi-Task Learning Framework for Session-based Recommendations](https://mlsanity.com/paper/2309.06533/) (2023)
- [DURENDAL: Graph deep learning framework for temporal heterogeneous networks](https://mlsanity.com/paper/2310.00336/) (2023)
- [GIMIRec: Global Interaction Information Aware Multi-Interest Framework for Sequential Recommendation](https://mlsanity.com/paper/2112.08717/) (2021)
- [BBE-LSWCM: A Bootstrapped Ensemble of Long and Short Window Clickstream Models](https://mlsanity.com/paper/2203.16155/) (2022)
- [Situating Recommender Systems in Practice: Towards Inductive Learning and Incremental Updates](https://mlsanity.com/paper/2211.06365/) (2022)
- [Can Transformers Predict Vibrations?](https://mlsanity.com/paper/2402.10511/) (2024)
- [A blockchain-based intelligent recommender system framework for enhancing supply chain resilience](https://mlsanity.com/paper/2404.00306/) (2024)
- [Robust OOD Graph Learning via Mean Constraints and Noise Reduction](https://mlsanity.com/paper/2506.19281/) (2025)
- [Sequential Learner Modeling Using Multi-Relational Graph Convolutional Networks](https://mlsanity.com/paper/2607.19253/) (2026)
- [Multi-Granularity Position Embedding of Graphs via Granular-Ball for Link Prediction](https://mlsanity.com/paper/2607.29115/) (2026)

## Abstract (arXiv)

Recommender systems that can learn from cross-session data to dynamically predict the next item a user will choose are crucial for online platforms. However, existing approaches often use out-of-the-box sequence models which are limited by speed and memory consumption, are often infeasible for production environments, and usually do not incorporate cross-session information, which is crucial for effective recommendations. Here we propose Hierarchical Temporal Convolutional Networks (HierTCN), a hierarchical deep learning architecture that makes dynamic recommendations based on users' sequential multi-session interactions with items. HierTCN is designed for web-scale systems with billions of items and hundreds of millions of users. It consists of two levels of models: The high-level model uses Recurrent Neural Networks (RNN) to aggregate users' evolving long-term interests across different sessions, while the low-level model is implemented with Temporal Convolutional Networks (TCN), utilizing both the long-term interests and the short-term interactions within sessions to predict the next interaction. We conduct extensive experiments on a public XING dataset and a large-scale Pinterest dataset that contains 6 million users with 1.6 billion interactions. We show that HierTCN is 2.5x faster than RNN-based models and uses 90% less data memory compared to TCN-based models. We further develop an effective data caching scheme and a queue-based mini-batch generator, enabling our model to be trained within 24 hours on a single GPU. Our model consistently outperforms state-of-the-art dynamic recommendation methods, with up to 18% improvement in recall and 10% in mean reciprocal rank.