用草图正则化实现科学模拟的实时神经压缩，防止遗忘且效果接近离线训练。
In Situ Training of Implicit Neural Compressors for Scientific Simulations via Sketch-Based Regularization
- 通过草图数据缓解持续学习中的灾难性遗忘问题。
- 在高压缩率下保持优异重建性能，长期模拟仍稳定有效。
- 适合需要实时压缩的科学仿真场景，如复杂几何与非结构网格。
针对隐式神经表征，本文提出一种新型原位训练协议，利用有限内存中的完整数据与草图数据样本，其中草图数据用于防止灾难性遗忘。理论层面，基于简单的Johnson-Lindenstrauss启发结果，解释了草图作为正则化器的有效性。虽然该方法对持续学习领域具有广泛意义，但本文重点应用于基于隐式神经表征的超网络进行原位神经压缩。我们在二维和三维复杂模拟数据上进行了评估，覆盖长时间跨度、非结构化网格及非笛卡尔几何。实验表明，在高压缩率下仍具备出色的重建性能；更重要的是，草图正则化使该原位方案近似达到等效离线方法的性能水平。
原文摘要 · Abstract (English)
Focusing on implicit neural representations, we present a novel in situ training protocol that employs limited memory buffers of full and sketched data samples, where the sketched data are leveraged to prevent catastrophic forgetting. The theoretical motivation for our use of sketching as a regularizer is presented via a simple Johnson-Lindenstrauss-informed result. While our methods may be of wider interest in the field of continual learning, we specifically target in situ neural compression using implicit neural representation-based hypernetworks. We evaluate our method on a variety of complex simulation data in two and three dimensions, over long time horizons, and across unstructured grids and non-Cartesian geometries. On these tasks, we show strong reconstruction performance at high compression rates. Most importantly, we demonstrate that sketching enables the presented in situ scheme to approximately match the performance of the equivalent offline method.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。