mlsanityData to 7 Oct 2026

SmaAT-QMix-UNet: A Parameter-Efficient Vector-Quantized UNet for Precipitation Nowcasting

arXiv 2603.21879 ↗First posted 23 Mar 2026cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Weather forecasting supports critical socioeconomic activities and complements environmental protection, yet operational Numerical Weather Prediction (NWP) systems remain computationally intensive, thus being inefficient for certain applications. Meanwhile, recent advances in deep data-driven models have demonstrated promising results in nowcasting tasks. This paper presents SmaAT-QMix-UNet, an enhanced variant of SmaAT-UNet that introduces two key innovations: a vector quantization (VQ) bottleneck at the encoder-decoder bridge, and mixed kernel depth-wise convolutions (MixConv) replacing selected encoder and decoder blocks. These enhancements both reduce the model's size and improve its nowcasting performance. We train and evaluate SmaAT-QMix-UNet on a Dutch radar precipitation dataset (2016-2019), predicting precipitation 30 minutes ahead. Three configurations are benchmarked: using only VQ, only MixConv, and the full SmaAT-QMix-UNet. Grad-CAM saliency maps highlight the regions influencing each nowcast, while a UMAP embedding of the codewords illustrates how the VQ layer clusters encoder outputs. The source code for SmaAT-QMix-UNet is publicly available on GitHub: https://github.com/nstavr04/MasterThesisSnellius.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 25 of its 30 references that are in the map. First posted on arXiv 23 Mar 2026.

Builds on

25 of 30 references.

References · 25

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  2. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  3. 2017Neural Discrete Representation Learningcited by 4,650
  4. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  5. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,841
  6. 2015Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcastingcited by 1,428
  7. 2021VideoGPT: Video Generation using VQ-VAE and Transformerscited by 651
  8. 2019GhostNet: More Features from Cheap Operationscited by 410
  9. 2017Deep Learning for Precipitation Nowcasting: A Benchmark and A New Modelcited by 180
  10. 2020MetNet: A Neural Weather Model for Precipitation Forecastingcited by 128
  11. 2018PredRNN++: Towards A Resolution of the Deep-in-Time Dilemma in Spatiotemporal Predictive Learningcited by 122
  12. 2019MixConv: Mixed Depthwise Convolutional Kernelscited by 108
13 more
  1. 2020SmaAt-UNet: Precipitation Nowcasting using a Small Attention-UNet Architecturecited by 71
  2. 2021Broad-UNet: Multi-scale feature learning for nowcasting taskscited by 25
  3. 2022AA-TransUNet: Attention Augmented TransUNet For Nowcasting Taskscited by 22
  4. 2021Deep Graph Convolutional Networks for Wind Speed Predictioncited by 19
  5. 2020Wind speed prediction using multidimensional convolutional neural networkscited by 17
  6. 2021Symbolic regression for scientific discovery: an application to wind speed forecastingcited by 16
  7. 2021Multistream Graph Attention Networks for Wind Speed Forecastingcited by 12
  8. 2024GA-SmaAt-GNet: Generative Adversarial Small Attention GNet for Extreme Precipitation Nowcastingcited by 9
  9. 2022Vector Quantisation for Robust Segmentationcited by 8
  10. 2024GPTCast: a weather language model for precipitation nowcastingcited by 8
  11. 2024Graph Dual-stream Convolutional Attention Fusion for Precipitation Nowcastingcited by 7
  12. 2023SAR-UNet: Small Attention Residual UNet for Explainable Nowcasting Taskscited by 6
  13. 2024VQUNet: Vector Quantization U-Net for Defending Adversarial Atacks by Regularizing Unwanted Noisecited by 2

Built on by

Not cited yet.