Artificial Intelligence1 min read

New approach enables physically consistent AI models of chaotic systems

By · Published by Everything Blog

In short

Predicting chaotic systems is notoriously difficult because minor initial variations can cause long-term machine learning forecasts to explode toward infinite, unrealistic values. Previous techniques used training penalties to discourage runaway behavior, but they lacked theoretical guarantees once exposed to unfamiliar real-world conditions. Published in the Proceedings of the National Academy of Sciences, the MIT team's method establishes formal restrictions that prevent unbounded drift while maintaining dynamic flexibility. Researchers are making the underlying information freely available to support broader applications in hydrology, climate science, and industrial systems.

Key points

  • The framework marks the first work to establish formal guarantees that machine learning m…: The framework marks the first work to establish formal guarantees that machine learning models of chaotic systems will stay bounded.
  • The approach prevents artificial runaway behavior during extended simulations without sac…: The approach prevents artificial runaway behavior during extended simulations without sacrificing the flexibility needed for nonlinear dynamics.
  • The technique aims to substantially improve predictive reliability across weather forecas…: The technique aims to substantially improve predictive reliability across weather forecasting, climate modeling, hydrology, and aeronautics.

Sources checked by our analysis pipeline: 1

Modeling of complex, chaotic processes such as weather and climate patterns and turbulent flows in aviation or in ocean currents has advanced enormously in recent years. However, such models are still prone to drift into physically impossible predictions over time.

Now, researchers with MIT’s Laboratory for Information and Decision Systems (LIDS) have found a way to introduce physics-based restrictions on machine learning models that can ensure that such models never exceed realistic bounds, based on what is actually possible in the real world. This new approach, using information that is now being made freely available by the team, could significantly improve the usefulness and reliability of such models, which besides climate and weather are also used in a wide range of industrial processes, aeronautics, and hydrology.

The new findings are published in the Proceedings of the National Academy of Sciences (PNAS) in a paper by MIT graduate student Sunbochen Tang and MIT professors Navid Azizan and Themistoklis Sapsis.

“Lack of guarantees for a bounded response has always been a major obstacle on formulating data-driven models for complex nonlinear systems,” says…

Artificial Intelligence