![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
ИИ против физики
Nov. 16th, 2023 01:28 pmВ журнале Nature опубликовали статью где сообщается, что DeepMind AI предсказывает погоду быстрее и лучше чем традиционные модели.
The model, called GraphCast, can run from a desktop computer and makes more accurate predictions than conventional models in minutes rather than hours.
AI models run 1,000 to 10,000 times faster than conventional numerical weather prediction (NWP) models
GraphCast, developed by Google’s AI company DeepMind in London, outperforms conventional and AI-based approaches at most global weather-forecasting tasks. Researchers first trained the model using estimates of past global weather made from 1979 to 2017 by physical models. This allowed GraphCast to learn links between weather variables such as air pressure, wind, temperature and humidity.
The trained model uses the ‘current’ state of global weather and weather estimates from 6 hours earlier to predict the weather 6 hours ahead. Earlier predictions are fed back into the model, enabling it to make estimates further into the future. DeepMind researchers found that GraphCast could use global weather estimates from 2018 to make forecasts up to 10 days ahead in less than a minute, and the predictions were more accurate than the ECMWF’s High RESolution forecasting system (HRES) — one version of its NWP — which takes hours to forecast.
Более подробная публикация в журнале Science.
GraphCast is implemented as a neural network architecture, based on GNNs in an “encode-process-decode” configuration, with a total of 36.7 million parameters
The multi-mesh contains the 40,962 nodes from the highest resolution mesh (which is roughly 1/25 the number of latitude/longitude grid points at 0.25°), and the union of all the edges created in the intermediate graphs, forming a flat hierarchy of edges with varying lengths.
Не помню кто мне рассказал как работали синоптики до внедрения ЭВМ. Они листали старые метеорологические карты чтобы найти ситуацию похожую на текущую, смотрели что было дальше и делали на основании этого прогноз.
The model, called GraphCast, can run from a desktop computer and makes more accurate predictions than conventional models in minutes rather than hours.
AI models run 1,000 to 10,000 times faster than conventional numerical weather prediction (NWP) models
GraphCast, developed by Google’s AI company DeepMind in London, outperforms conventional and AI-based approaches at most global weather-forecasting tasks. Researchers first trained the model using estimates of past global weather made from 1979 to 2017 by physical models. This allowed GraphCast to learn links between weather variables such as air pressure, wind, temperature and humidity.
The trained model uses the ‘current’ state of global weather and weather estimates from 6 hours earlier to predict the weather 6 hours ahead. Earlier predictions are fed back into the model, enabling it to make estimates further into the future. DeepMind researchers found that GraphCast could use global weather estimates from 2018 to make forecasts up to 10 days ahead in less than a minute, and the predictions were more accurate than the ECMWF’s High RESolution forecasting system (HRES) — one version of its NWP — which takes hours to forecast.
Более подробная публикация в журнале Science.
GraphCast is implemented as a neural network architecture, based on GNNs in an “encode-process-decode” configuration, with a total of 36.7 million parameters
The multi-mesh contains the 40,962 nodes from the highest resolution mesh (which is roughly 1/25 the number of latitude/longitude grid points at 0.25°), and the union of all the edges created in the intermediate graphs, forming a flat hierarchy of edges with varying lengths.
Не помню кто мне рассказал как работали синоптики до внедрения ЭВМ. Они листали старые метеорологические карты чтобы найти ситуацию похожую на текущую, смотрели что было дальше и делали на основании этого прогноз.
