# A Memory-Network Based Solution for Multivariate Time-Series Forecasting

arXiv: https://arxiv.org/abs/1809.02105 · HTML: https://mlsanity.com/paper/1809.02105/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Sep 2018.

## Built on by (most cited first, 22 of 22)

- [Flash-VStream: Memory-Based Real-Time Understanding for Long Video Streams](https://mlsanity.com/paper/2406.08085/) (2024)
- [Flash-VStream: Efficient Real-Time Understanding for Long Video Streams](https://mlsanity.com/paper/2506.23825/) (2025)
- [Fredformer: Frequency Debiased Transformer for Time Series Forecasting](https://mlsanity.com/paper/2406.09009/) (2024)
- [Multi-Scale Adaptive Graph Neural Network for Multivariate Time Series Forecasting](https://mlsanity.com/paper/2201.04828/) (2022)
- [Mixture-of-Linear-Experts for Long-term Time Series Forecasting](https://mlsanity.com/paper/2312.06786/) (2023)
- [A Multi-view Multi-task Learning Framework for Multi-variate Time Series Forecasting](https://mlsanity.com/paper/2109.01657/) (2021)
- [LightCTS: A Lightweight Framework for Correlated Time Series Forecasting](https://mlsanity.com/paper/2302.11974/) (2023)
- [RNN with Particle Flow for Probabilistic Spatio-temporal Forecasting](https://mlsanity.com/paper/2106.06064/) (2021)
- [A Survey of Explainable Artificial Intelligence (XAI) in Financial Time Series Forecasting](https://mlsanity.com/paper/2407.15909/) (2024)
- [From Known to Unknown: Knowledge-guided Transformer for Time-Series Sales Forecasting in Alibaba](https://mlsanity.com/paper/2109.08381/) (2021)
- [A Systematic Review for Transformer-based Long-term Series Forecasting](https://mlsanity.com/paper/2310.20218/) (2023)
- [Scale-Aware Neural Architecture Search for Multivariate Time Series Forecasting](https://mlsanity.com/paper/2112.07459/) (2021)
- [Multimodal Quasi-AutoRegression: Forecasting the visual popularity of new fashion products](https://mlsanity.com/paper/2204.04014/) (2022)
- [Learning Heterogeneous Interaction Strengths by Trajectory Prediction with Graph Neural Network](https://mlsanity.com/paper/2208.13179/) (2022)
- [Dynamical Deep Generative Latent Modeling of 3D Skeletal Motion](https://mlsanity.com/paper/2106.10393/) (2021)
- [Selecting Datasets for Evaluating an Enhanced Deep Learning Framework](https://mlsanity.com/paper/2109.10442/) (2021)
- [Sub-Graph Learning for Spatiotemporal Forecasting via Knowledge Distillation](https://mlsanity.com/paper/2211.09740/) (2022)
- [Pattern-Matching Dynamic Memory Network for Dual-Mode Traffic Prediction](https://mlsanity.com/paper/2408.07100/) (2024)
- [Mobility-based Traffic Forecasting in a Multimodal Transport System](https://mlsanity.com/paper/2411.08052/) (2024)
- [Forecast-Then-Optimize Deep Learning Methods](https://mlsanity.com/paper/2506.13036/) (2025)
- [DB2-TransF: All You Need Is Learnable Daubechies Wavelets for Time Series Forecasting](https://mlsanity.com/paper/2512.10051/) (2025)
- [Memory in Deep Time-Series Models](https://mlsanity.com/paper/2609.06006/) (2026)

## Abstract (arXiv)

Multivariate time series forecasting is extensively studied throughout the years with ubiquitous applications in areas such as finance, traffic, environment, etc. Still, concerns have been raised on traditional methods for incapable of modeling complex patterns or dependencies lying in real word data. To address such concerns, various deep learning models, mainly Recurrent Neural Network (RNN) based methods, are proposed. Nevertheless, capturing extremely long-term patterns while effectively incorporating information from other variables remains a challenge for time-series forecasting. Furthermore, lack-of-explainability remains one serious drawback for deep neural network models. Inspired by Memory Network proposed for solving the question-answering task, we propose a deep learning based model named Memory Time-series network (MTNet) for time series forecasting. MTNet consists of a large memory component, three separate encoders, and an autoregressive component to train jointly. Additionally, the attention mechanism designed enable MTNet to be highly interpretable. We can easily tell which part of the historic data is referenced the most.