Nvidia's latest AI climate model takes aim at severe weather

While enterprises struggle to quantify the return on investment of AI, the technology continues to show promise in bolstering weather forecasting and climate models. On Monday, Nvidia unveiled a new generative AI diffusion model, developed in collaboration with Lawrence Berkeley National Lab and the University of Washington, …

[HuBo]

Stormy Daniels

Nicely illustrated at the top of the Nvidia Blog Post (under the "unveiled" link, 3rd line of TFA). And, expected to be economical as well (from the Blog): "lifesaving work, which previously cost nearly $3 million on CPUs, can be accomplished using about $60,000 on a single system with an NVIDIA H100 Tensor Core GPU" -- (or from Tobias) "far less costly to run than CPU-based compute clusters" -- if it works reliably of course, in Taiwan and elsewhere.

The Blog links to the Stormcast 150MB pdf preprint (lotsa figures in appendices), and so you know what I'm thinking, right? Well, here it is: Should generative AI's stable diffusion be expected to be capable of accurately predicting the unstable nonlinear phenomena observed in the atmosphere? Wouldn't "unstable diffusion" be preferrable there (if it exists)?

In the classical k-ε model, the eddy viscosity (μt ∝ k²/ε) gets singularly infinite where the rate of dissipation of turbulent kinetic energy (ε) somehow vanishes, resulting in realistic flowfield mayhem. Might genAI need unstable diffusion to produce a similar walk on the wild side? (just a thought -- not an expert ...)

[Spherical Cow]

Re: Stormy Daniels

Can you select all images which contain fire hydrants?

[elsergiovolador]

Weathermarket

If they can forecast weather, they can forecast the markets.

[AVR]

Re: Weathermarket

Meteorologists do forecast the weather, and almost without exception aren't rich. Not with 100% reliability, but enough to be useful.

[Anonymous Coward]

Re: Weathermarket

Stock market crystal ball fortune telling is AI's killer app really ... like BloombergGPT IMHO. Turbulence and chaos ahoy ...

[LionelB]

Re: Weathermarket

Atmospheric dynamics at least are underpinned by (mostly) known, if inherently difficult-to-predict, deterministic (if chaotic) physics. Market dynamics are underpinned by human individual and herd psychology in all its perverse and irrational glory, and (as I learned during a brief stint as a quant), a woefully poor signal-to-noise ratio. Those are very different beasts.

[MONK_DUCK]

Re: Weathermarket

Of course they can, in fact they already are doing it, and once several finace houses are doing the at scale the benefits of it will be gone and we will just go back to looking at the fundamentals.

[Korev]

Google shared its work with the European Centre for Medium-Range Weather Forecasts (ECMWF) to not only enhance physics-based climate models with machine learning, but port them to TPUs and GPUs, making it far less costly to run than CPU-based compute clusters.

Not exactly the same thing, but Meteoswiss & CSCS ported COSMO to GPUs a few years ago too. Since then, Swiss short range forecasts have been made on a pair of surprisingly small GPU clusters.