arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.05288cs.LGcs.AIcs.CEcs.CV
收藏

内部变率对深度学习气候仿真器基准测试的影响

The impact of internal variability on benchmarking deep learning climate emulators

Björn Lütjens, Raffaele Ferrari, Duncan Watson-Parris, Noelle Selin

AI总结:

本文研究ClimateBench中内部变率对深度学习气候仿真器评估的影响,发现低频噪声会导致过拟合,并通过增加MPI-ESM1.2-LR集合模拟改进目标。

摘要:

全复杂度地球系统模型(ESM)在计算上非常昂贵,限制了其用于探索多种排放路径的气候结果。更高效的、近似ESM的仿真器可以直接将排放映射到气候结果,而基准测试正被用于在标准化任务和数据集上评估其准确性。我们研究了数据驱动气候仿真中一个流行的基准ClimateBench,目前基于深度学习的仿真器在该基准上取得了最佳性能。我们将这些深度学习仿真器与一个类似于模式缩放的、基于线性回归的仿真器进行比较,并表明它在4个区域分辨气候变量中的3个上优于现有1亿参数深度学习基础模型ClimaX,尤其是地表温度和降水。虽然仿真地表温度预计主要是线性的,但这一结果对于降水仿真而言令人惊讶。降水是一个噪声大得多的变量,我们表明深度学习仿真器可能会过拟合低频的内部变率噪声,从而相对于线性仿真器降低其性能。我们通过增加每条排放路径的气候模拟数量(从3个增加到50个),并用MPI-ESM1.2-LR模型相应的集合平均值更新基准目标,来解决过拟合问题。使用新目标后,我们表明线性模式缩放在温度上仍然更准确,但在仿真降水时可以被一种基于深度学习的技术超越。我们在github.com/blutjens/climate-emulator发布代码和数据。

英文摘要:

Full-complexity Earth system models (ESMs) are computationally very expensive, limiting their use in exploring the climate outcomes of multiple emission pathways. More efficient emulators that approximate ESMs can directly map emissions onto climate outcomes, and benchmarks are being used to evaluate their accuracy on standardized tasks and datasets. We investigate a popular benchmark in data-driven climate emulation, ClimateBench, on which deep learning-based emulators are currently achieving the best performance. We compare these deep learning emulators with a linear regression-based emulator, akin to pattern scaling, and show that it outperforms the incumbent 100M-parameter deep learning foundation model, ClimaX, on 3 out of 4 regionally-resolved climate variables, notably surface temperature and precipitation. While emulating surface temperature is expected to be predominantly linear, this result is surprising for emulating precipitation. Precipitation is a much more noisy variable, and we show that deep learning emulators can overfit to internal variability noise at low frequencies, degrading their performance in comparison to a linear emulator. We address the issue of overfitting by increasing the number of climate simulations per emission pathway (from 3 to 50) and updating the benchmark targets with the respective ensemble averages from the MPI-ESM1.2-LR model. Using the new targets, we show that linear pattern scaling continues to be more accurate on temperature, but can be outperformed by a deep learning-based technique for emulating precipitation. We publish our code and data at github.com/blutjens/climate-emulator.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑