Purely Data-Driven Approaches to Weather Prediction: Promise and Perils

  • * Register (or log in) to the Neural Network to add this session to your agenda or watch the replay

  • Date
    5 January 2022
    Timeframe
    17:00 - 18:30 CET, Geneva
    Duration
    90 minutes
    Share this session

    The use of machine learning (ML) models for weather prediction has emerged as a popular area of research. The promise of these models — whether in conjunction with more traditional Numerical Weather Prediction (NWP), or on its own — is that they allow for more accurate predictions of the weather at significantly reduced computational cost.

    This AI for Good Discovery featuring Suman Ravuri, Staff Research Scientist at DeepMind, will discuss both the promise and perils of using a data-driven (and more specifically deep learning) only approach. It will highlight a recent project with the Met Office on precipitation nowcasting as a case study. While the study found that we could create a deep learning model that was significantly preferred by Met Office meteorologists and performed well on objective measures of performance, we also discovered many ways in which deep learning systems can perform well on objective measures of performance without improving decision-making value. This talk will discuss some reasons why this failure mode occurs, while also advocating for better verification of purely data-driven models. Although the discussion focuses on very-short-term prediction, we believe that many of these lessons are also applicable to longer-term forecasts using machine learning for weather prediction.

  • Share this session
    Discover more from programme stream
    Machine learning and climate change: learning from present-day observations to...

    Machine learning and climate change: learning from present-day observations to...

    13 March 2024 - 17:00 to 18:15

    Climate Causality: Connecting data and theory to understand the drivers...

    Climate Causality: Connecting data and theory to understand the drivers...

    6 March 2024 - 17:00 to 18:15

    Creating trustworthy AI for weather and climate

    Creating trustworthy AI for weather and climate

    21 February 2024 - 17:00 to 18:15

    WeatherBench 2 and the second revolution of weather prediction

    WeatherBench 2 and the second revolution of weather prediction

    29 November 2023 - 17:00 to 18:15

    Machine Learning Insights into Precipitation: Implicitly Learning Cloud Organization for...

    Machine Learning Insights into Precipitation: Implicitly Learning Cloud Organization for...

    15 November 2023 - 14:00 to 15:15

    Are you sure you want to remove this speaker?