# Memory In Memory: A Predictive Neural Network for Learning Higher-Order Non-Stationarity from Spatiotemporal Dynamics

arXiv: https://arxiv.org/abs/1811.07490 · HTML: https://mlsanity.com/paper/1811.07490/

## Key facts

- Cited by 92 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Nov 2018.

## Built on by (most cited first, 30 of 92)

- [SimVP: Simpler yet Better Video Prediction](https://mlsanity.com/paper/2206.05099/) (2022)
- [PredRNN: A Recurrent Neural Network for Spatiotemporal Predictive Learning](https://mlsanity.com/paper/2103.09504/) (2021)
- [FitVid: Overfitting in Pixel-Level Video Prediction](https://mlsanity.com/paper/2106.13195/) (2021)
- [UniST: A Prompt-Empowered Universal Model for Urban Spatio-Temporal Prediction](https://mlsanity.com/paper/2402.11838/) (2024)
- [Temporal Attention Unit: Towards Efficient Spatiotemporal Predictive Learning](https://mlsanity.com/paper/2206.12126/) (2022)
- [Latency-Aware Collaborative Perception](https://mlsanity.com/paper/2207.08560/) (2022)
- [MotionRNN: A Flexible Model for Video Prediction with Spacetime-Varying Motions](https://mlsanity.com/paper/2103.02243/) (2021)
- [OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning](https://mlsanity.com/paper/2306.11249/) (2023)
- [PastNet: Introducing Physical Inductive Biases for Spatio-temporal Video Prediction](https://mlsanity.com/paper/2305.11421/) (2023)
- [VMRNN: Integrating Vision Mamba and LSTM for Efficient and Accurate Spatiotemporal Forecasting](https://mlsanity.com/paper/2403.16536/) (2024)
- [SwinLSTM:Improving Spatiotemporal Prediction Accuracy using Swin Transformer and LSTM](https://mlsanity.com/paper/2308.09891/) (2023)
- [SimVPv2: Towards Simple yet Powerful Spatiotemporal Predictive Learning](https://mlsanity.com/paper/2211.12509/) (2022)
- [MMVP: Motion-Matrix-based Video Prediction](https://mlsanity.com/paper/2308.16154/) (2023)
- [Dynamics-Aware Spatiotemporal Occupancy Prediction in Urban Environments](https://mlsanity.com/paper/2209.13172/) (2022)
- [MIMO Is All You Need : A Strong Multi-In-Multi-Out Baseline for Video Prediction](https://mlsanity.com/paper/2212.04655/) (2022)
- [Fourier Amplitude and Correlation Loss: Beyond Using L2 Loss for Skillful Precipitation Nowcasting](https://mlsanity.com/paper/2410.23159/) (2024)
- [Implicit Stacked Autoregressive Model for Video Prediction](https://mlsanity.com/paper/2303.07849/) (2023)
- [Deep learning for precipitation nowcasting: A survey from the perspective of time series forecasting](https://mlsanity.com/paper/2406.04867/) (2024)
- [TempEE: Temporal-Spatial Parallel Transformer for Radar Echo Extrapolation Beyond Auto-Regression](https://mlsanity.com/paper/2304.14131/) (2023)
- [Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory](https://mlsanity.com/paper/2308.13236/) (2023)
- [Diffusion Transformers as Open-World Spatiotemporal Foundation Models](https://mlsanity.com/paper/2411.12164/) (2024)
- [Conditional Temporal Variational AutoEncoder for Action Video Prediction](https://mlsanity.com/paper/2108.05658/) (2021)
- [Deterministic Guidance Diffusion Model for Probabilistic Weather Forecasting](https://mlsanity.com/paper/2312.02819/) (2023)
- [Modeling Spatio-temporal Dynamical Systems with Neural Discrete Learning and Levels-of-Experts](https://mlsanity.com/paper/2402.05970/) (2024)
- [Triplet Attention Transformer for Spatiotemporal Predictive Learning](https://mlsanity.com/paper/2310.18698/) (2023)
- [DynST: Dynamic Sparse Training for Resource-Constrained Spatio-Temporal Forecasting](https://mlsanity.com/paper/2403.02914/) (2024)
- [UniFlow: A Foundation Model for Unified Urban Spatio-Temporal Flow Prediction](https://mlsanity.com/paper/2411.12972/) (2024)
- [Video Prediction Transformers without Recurrence or Convolution](https://mlsanity.com/paper/2410.04733/) (2024)
- [MS-RNN: A Flexible Multi-Scale Framework for Spatiotemporal Predictive Learning](https://mlsanity.com/paper/2206.03010/) (2022)
- [S-HR-VQVAE: Sequential Hierarchical Residual Learning Vector Quantized Variational Autoencoder for Video Prediction](https://mlsanity.com/paper/2307.06701/) (2023)

## Abstract (arXiv)

Natural spatiotemporal processes can be highly non-stationary in many ways, e.g. the low-level non-stationarity such as spatial correlations or temporal dependencies of local pixel values; and the high-level variations such as the accumulation, deformation or dissipation of radar echoes in precipitation forecasting. From Cramer's Decomposition, any non-stationary process can be decomposed into deterministic, time-variant polynomials, plus a zero-mean stochastic term. By applying differencing operations appropriately, we may turn time-variant polynomials into a constant, making the deterministic component predictable. However, most previous recurrent neural networks for spatiotemporal prediction do not use the differential signals effectively, and their relatively simple state transition functions prevent them from learning too complicated variations in spacetime. We propose the Memory In Memory (MIM) networks and corresponding recurrent blocks for this purpose. The MIM blocks exploit the differential signals between adjacent recurrent states to model the non-stationary and approximately stationary properties in spatiotemporal dynamics with two cascaded, self-renewed memory modules. By stacking multiple MIM blocks, we could potentially handle higher-order non-stationarity. The MIM networks achieve the state-of-the-art results on four spatiotemporal prediction tasks across both synthetic and real-world datasets. We believe that the general idea of this work can be potentially applied to other time-series forecasting tasks.