HSE Researchers Join Forces with Yandex Cloud to Develop a Neural Network for Predicting El Niño
A team of researchers from HSE University, jointly with the Yandex School of Data Analysis and Yandex Cloud, have developed a neural network for anticipating El Niño climate anomalies. The new algorithm enables more precise predictions of changes in the average surface temperature of oceanic waters that can trigger natural disasters in specific regions of the world. At present, the model is capable of predicting El Niño events one and a half years in advance, and the researchers are working towards extending the forecast period to two years.
El Niño refers to a shift in the distribution of surface water temperatures in the Pacific Ocean, which impacts weather patterns and can potentially trigger natural disasters in certain regions.
El Niño leads to an unusual warming of the equatorial region, and the neural network is capable of simulating the future average temperatures in the Pacific equatorial zone. There is also a reverse process called La Niña, which involves a decrease in oceanic temperatures. The cycle of change between the two processes occurs at intervals of 2 to 7 years. These fluctuations have a significant impact on weather patterns in several countries across the globe and can elevate the risk of wildfires, droughts, floods, and crop failures.
The research team trained the neural network on a set of thousands of temperature maps using synthetic and real data collected from the year 1800 to the present day. Besides conventional machine learning techniques for predicting such phenomena, the ML specialists have been experimenting with the Autoformer architecture during the training process. This approach facilitates high-quality processing of a sequence of temperature maps. To pre-process the datasets, the researchers employed the Yandex DataSphere ML development service, which offers all the necessary tools and dynamically scalable cloud resources for the entire machine learning development cycle.
Dmitry Vetrov
'The challenges of global climate change are becoming increasingly urgent. What's truly alarming isn't just the warming itself, but rather the inevitable imbalance that our planet's climate will undergo. The El Niño effect plays a critical role in causing global weather and climate fluctuations, leading to significant adverse events such as crop failures. Therefore, anticipating El Niño is particularly vital in the current context of escalating climate imbalance,' says Dmitry Vetrov, Research Professor at the Big Data and Information Retrieval School, HSE Faculty of Computer Science, and leading researcher at AIRI.
Anna Lemyakina
'Cloud-based technologies assist us in conducting scientific experiments more efficiently. In projects like El Niño research, quick and flexible access to services for testing various machine learning models is critical. Each test using a new architecture aids in predicting the phenomenon as accurately and early as possible,' explains Anna Lemyakina, Director of National Strategic Projects at Yandex Cloud.
See also:
Neural Network Developed at HSE Campus in Perm Will Determine Root Cause of Stroke in Patients
Specialists at HSE Campus in Perm and clinicians at Perm City Clinical Hospital No. 4, have been collaborating to develop a neural network capable of determining the root cause of a stroke. This marks the world's first attempt to create such a system, the developers note.
HSE Researchers Teach Neural Networks to Better Detect Humour
A group of scientists from the HSE Faculty of Computer Science has conducted a study on the ability of neural networks to detect humour. It turns out that for more reliable recognition, it’s necessary to change the approach to creating datasets on which neural networks are trained. The scientists presented these results at one of the world's most important conferences on natural language processing — EMNLP 2023.
Neural Networks of Power: AI Unravels Knots and Tangles in Relationships between Humans, Elves and Hobbits
One of the most popular writers of the last century, John Ronald Reuel Tolkien, was born on January 3rd. Researchers from HSE University, AIRI and MISSIS have used machine learning to explore the social connections between the characters of his Middle-earth universe. The algorithm managed to create an accurate picture of the social structures and dynamics of the characters' relationships, providing a unique map of interactions in the epic world. The results of the work were published in IEEE Xplore.
Specialists from the HSE Institute of Education Confirm GigaChat’s Erudition in Social Sciences
A multimodal neural network model by Sber, under the supervision of HSE University’s expert commission, has successfully passed the Unified State Exam in social studies. GigaChat completed all exam tasks and scored 67 points.
Child Ex Machina: What Artificial Intelligence Can Learn from Toddlers
Top development teams around the world are trying to create a neural network similar to a curious but bored three-year-old kid. IQ.HSE shares why this approach is necessary and how such methods can bring us closer to creating strong artificial intelligence.
‘Neural Networks Are Something that Move the World Forward’
On September 4, the HSE University building on Pokrovsky Bulvar hosted ARTificial Fest, an event devoted to neural network art. The festival was organised by the HSE University Faculty of Creative Industries, the HSE Career centre, and the Chisty List (‘Blank Page’) student organisation. The event was open not only to students and staff of HSE University, but also to anyone interested in the blending of machine algorithms and art.
Faster and More Precise: Researcher Improves Performance of Image Recognition Neural Network
A scientist from HSE University has developed an image recognition algorithm that works 40% faster than analogues. It can speed up real-time processing of video-based image recognition systems. The results of the study have been published in the journal Information Sciences.
Researchers Teach Neural Networks to Recognize Similar Objects on Videos without Accuracy Degradation
Andrey Savchenko, Professor at HSE University, has developed a method that can help to enhance image identification on videos. In his project, a network was taught by a new algorithm and can now make decisions on image recognition and classification 10 times faster than before.