物理沒有基礎模型，神經算子用天氣預測補上這一課

核心論點 · 點時間戳可跳到原聲

原話 · 已逐字校驗

we can now have neural networks be part of the verification loop and have confidence that we can use them appropriately

我們現在能讓神經網絡成為驗證迴路的一部分，並確信可以恰當地使用它們。

Anima Anandkumar6:07

we're talking like hundreds of billions to even a trillion context length right so forget ever having a transformer for anything of this scale all of the world's compute will not be enough

這意味著數百億甚至萬億的上下文長度，所以別指望 Transformer 能處理這種規模——全世界的算力加起來都不夠。

Anima Anandkumar29:25

it's not only you know accurate it's almost as close to the uh what the traditional weather models can do accurately but also tens of thousands of times faster

它不僅準確，幾乎能逼近傳統天氣模型能做到的精度，而且快數萬倍。

Anima Anandkumar33:29

there's only one earth. You know, if you want a foundation model, if the claim is that it should be able to do both very short-term as well as very long-term together.

地球只有一個。如果你想做基礎模型，那它就應該能同時處理極短期和極長期預測。

Anima Anandkumar36:30

the physical world may be more forgiving because you know where there are extreme events like hurricanes that have very specific physical signature

物理世界可能更寬容，因為像颶風這樣的極端事件帶有非常具體的物理特徵。

Anima Anandkumar43:37

we have foundation models for language maybe vision but not for physics

我們有語言的基礎模型，也許還有視覺的，但沒有物理的。

Anima Anandkumar1:12:55

without that we cannot experiment, we cannot innovate. I think this is a part that I push a lot

沒有它，我們就無法實驗、無法創新。這是我一直非常強調的一點。

Anima Anandkumar1:21:01

數字與實體

術語

Neural Operator神經算子

學習函數空間之間映射的神經網絡，輸入輸出可任意分辨率，適合物理場建模。

Fourier Neural Operator傅里葉神經算子

在傅里葉域建模全局依賴的神經算子，複雜度近線性，兼顧效率與表達力。

Physics-informed neural networks物理信息神經網絡

把 PDE 作為損失約束的神經網絡；簡單問題有效，時間依賴和湍流問題容易失敗。

Certified Robustness認證魯棒性

用形式化方法給出輸入擾動下輸出變化的上界，用於安全關鍵場景。

Reanalysis data再分析資料

把衛星觀測與物理模式同化後的歷史氣象數據，作為 AI 天氣模型的訓練集。

Ensemble prediction集合預報

對初始條件加擾動跑多條軌跡，再統計得到概率化的預報結果。

收聽指南