Google (NASDAQ: GOOGL) is planting its feet deeper in artificial intelligence (AI) with its research arm confirming the launch of an AI-based weather forecasting model capable of predicting weather changes that are difficult to spot.
Dubbed the Scalable Ensemble Envelope Diffusion Sampler (SEEDS), Google’s AI-based model bears striking similarities with mainstream large language models (LLMs) and other diffusion models.
According to a paper published in Science Advances, SEEDS can generate ensembles of weather forecasts at scale, outstripping the capabilities of traditional forecasting systems. The AI system leverages probabilistic diffusion models akin to image and video generators like Stable Diffusion and Midjourney.
“We present SEEDS, [a] new AI technology to accelerate and improve weather forecasts using diffusion models,” read the announcement. “SEEDS enables a significant reduction in computational cost for generating ensemble forecasts and better characterization of rare or extreme weather events.”
SEEDS stands apart thanks to Google’s state-of-the-art denoising diffusion probabilistic models, which allow it to generate realistic weather predictions. Per the research paper, SEEDS only requires one forecast from a reputable numerical weather prediction system to develop its large pool of predictions.
When placed side-by-side with physics-based weather prediction systems, SEEDS predictions appear superior upon measurements using rank histogram, continuous ranked probability score (CRPS), and root-mean-squared error (RMSE).
Apart from its superior output, the computational cost of the model is incomparable with traditional models, with the report describing it as “negligible.” Google Research indicates that SEEDS is ahead of its peers when covering extreme events such as heat waves while offering the perks of scalability.
“Specifically, our highly scalable generative approach enables the creation of very large ensembles that can characterize very rare events by providing samples of weather states exceeding a given threshold for any user-defined diagnostic,” read the report.
Using technology to save the planet
Since AI became mainstream, several environmental conservationists have turned to the technology to advance their goals of saving the planet. Researchers from Johns Hopkins and the National Oceanic and Atmospheric Administration (NOAA) are using AI models to predict weather patterns to stifle the impacts of pollution.
India is also following the same path with its meteorological department keen on leveraging emerging technologies to predict weather events like droughts and flash floods. Armed with recent innovations, Australia-based charity ClimateForce says it will use AI to preserve the ecological balance of the Daintree rainforest in partnership with the NTT Group.
In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.
Watch IEEE COINS Conference: Intersection of blockchain, AI, IoT & IPv6 technologies
New to blockchain? Check out CoinGeek’s Blockchain for Beginners section, the ultimate resource guide to learn more about blockchain technology.
