News from the AI & ML world
DeeperML - #aiweather
|
Sana Hassan@MarkTechPost
//
Google has unveiled a new AI model designed to forecast tropical cyclones with improved accuracy. Developed through a collaboration between Google Research and DeepMind, the model is accessible via a newly launched website called Weather Lab. The AI aims to predict both the path and intensity of cyclones days in advance, overcoming limitations present in traditional physics-based weather prediction models. Google claims its algorithm achieves "state-of-the-art accuracy" in forecasting cyclone track and intensity, as well as details like formation, size, and shape.
The AI model was trained using two extensive datasets: one describing the characteristics of nearly 5,000 cyclones from the past 45 years, and another containing millions of weather observations. Internal testing demonstrated the algorithm's ability to accurately predict the paths of recent cyclones, in some cases up to a week in advance. The model can generate 50 possible scenarios, extending forecast capabilities up to 15 days. This breakthrough has already seen adoption by the U.S. National Hurricane Center, which is now using these experimental AI predictions alongside traditional forecasting models in its operational workflow. Google's AI's ability to forecast up to 15 days in advance marks a significant improvement over current models, which typically provide 3-5 day forecasts. Google made the AI accessible through a new website called Weather Lab. The model is available alongside two years' worth of historical forecasts, as well as data from traditional physics-based weather prediction algorithms. According to Google, this could help weather agencies and emergency service experts better anticipate a cyclone’s path and intensity. Recommended read:
References :
Source Asia@Source Asia
//
Microsoft's Aurora AI model is revolutionizing weather forecasting by providing accurate 10-day forecasts in mere seconds. This AI foundation model, developed by Microsoft Research, has demonstrated capabilities that extend beyond traditional weather prediction, encompassing environmental events such as tropical cyclones, air quality, and ocean waves. Aurora achieves this by training on a massive dataset of over one million hours of atmospheric data from satellites, radar, weather stations, simulations, and forecasts, which Microsoft believes is the largest collection ever assembled for training an AI forecasting model. The model's speed and accuracy have the potential to improve safety and inform decisions across various sectors.
The core strength of Aurora lies in its foundation model architecture. It's not simply limited to weather forecasting; it can be fine-tuned for specific environmental prediction tasks. After initial training on general weather patterns, Aurora can be adapted with smaller datasets to forecast elements like wave height or air quality. The AI does not fully grasp the physical laws governing weather, but its use for environmental prediction tasks and ability to provide accurate forecasts is still significant. This flexibility makes it a versatile tool for understanding and predicting various aspects of the Earth system. Aurora's performance has been noteworthy, beating existing numerical and AI models across 91 percent of forecasting targets when fine-tuned to medium-range weather forecasts. Its rapid processing time, taking seconds compared to the hours required by traditional models, makes it a valuable asset for timely decision-making. Microsoft is leveraging AI technology to make weather forecasting more efficient and accurate. While generative AI is revolutionizing how we do things, integrating it into workflows is making work easier by automating redundant tasks, creating more time to focus on more important tasks. Recommended read:
References :
Source Asia@Source Asia
//
Microsoft's Aurora AI foundation model is revolutionizing weather and environmental forecasting, offering quicker and more accurate predictions compared to traditional methods. Developed by Microsoft Research, Aurora is a large-scale AI model trained on a vast dataset of atmospheric information, including satellite data, radar readings, weather station observations, and simulations. This comprehensive training allows Aurora to forecast a range of environmental events, from hurricanes and typhoons to air quality and ocean waves, with exceptional precision and speed. The model's capabilities extend beyond conventional weather forecasting, making it a versatile tool for understanding and predicting environmental changes.
Aurora's unique architecture enables it to be fine-tuned for specific tasks using modest amounts of additional data. This "fine-tuning" process allows the model to generate forecasts in seconds, demonstrating its efficiency and adaptability. Researchers have shown that Aurora outperforms existing numerical and AI models in 91% of forecasting targets when fine-tuned for medium-range weather forecasts. Its ability to accurately predict hurricane trajectories and other extreme weather events highlights its potential to improve disaster preparedness and response efforts, ultimately saving lives and mitigating damage. Senior researchers Megan Stanley and Wessel Bruinsma emphasized Aurora's broader impact on environmental science, noting its potential to revolutionize the field. In a paper published in Nature, they highlighted Aurora's ability to correctly forecast hurricanes in 2023 more accurately than operational forecasting centers, such as the US National Hurricane Center. Aurora also demonstrated its capabilities when correctly forecasting where and when Doksuri would hit the Philippines four days in advance. These findings underscore the transformative potential of AI in addressing complex environmental challenges and paving the way for more effective climate modeling and environmental event management. Recommended read:
References :
Keshav Kumaresan@DagsHub Blog
//
References:
DagsHub Blog
, The Cognitive Revolution
,
AI is making waves in unexpected areas. A recent study has found that AI-generated memes are, on average, funnier and more shareable than those created solely by humans. Researchers from KTH Royal Institute of Technology, LMU Munich, and TU Darmstadt, discovered that memes crafted entirely by OpenAI's GPT-4 scored higher in humor, creativity, and shareability. However, human-created memes still hold the crown for the absolute funniest individual examples, showcasing the unique personal touch humans bring to humor.
The Cognitive Revolution podcast recently featured Andreessen Horowitz partners Olivia Moore and Anish Acharya discussing the rapid advancements in voice AI. The discussion explored how the latest improvements are enabling more natural voice interactions across various platforms. Businesses are already utilizing voice AI for tasks ranging from complex negotiations to after-hours customer support. Recommended read:
References :
|
BenchmarksBlogsResearch Tools |