The Pulse
MIT and UBC Researchers Develop WorldParticle to Simulate Phenomena from Sliding Sand to Folding Proteins
WorldParticle represents materials as particles and predicts their movement one step at a time. Researchers demonstrated the system on six different physical phenomena, including deformable objects, granular materials and proteins, and say
AI.info Team · Written with AI · fact-checked ·
Six phenomena, from sand to proteins
The researchers demonstrated WorldParticle by modeling six different physical phenomena, including deformable objects, granular materials and proteins. For instance, their method effectively captured how a granular object would collapse into a pile while some baseline methods left sand particles floating in the air. Their method also modeled how protein structures change over time about 100 times faster than numerical simulators, according to co-lead author Minghao Guo, an MIT postdoc.
WorldParticle represents materials as particles, from a dune of sand to a protein molecule
Researchers at MIT, the University of British Columbia (UBC) and elsewhere have developed a single deep-learning system called WorldParticle that could help engineers model different physical phenomena, from sliding sand to folding proteins, with one unified artificial intelligence architecture. It represents every material it simulates, from a dune of sand to a protein molecule, as tiny particles and predicts where those particles will move one step at a time. The researchers designed a neural network that learns to solve governing equations in general, instead of building a new model and solver for each material and process.
“It is a similar architecture to ChatGPT. But instead of predicting one word at a time to write a piece of text, we are predicting how these particles will move, one step at a time,”
Peter Yichen Chen, a former MIT postdoc who is now an assistant professor at UBC, and is a co-corresponding author of an open-access paper on this technique
WorldParticle separates external forces and compresses particles
Before each simulation step, WorldParticle separates the external forces that affect the physical system. After predicting the initial states of the particles, it integrates the external forces back into the simulation, correcting its particle predictions based on how those forces affect the object's behavior. To streamline processing, the researchers employed a merging technique to combine and compress particles while still allowing WorldParticle to capture fine-grained details of their behavior. The system can adapt to materials, forces and environments it never encountered during training.
From soft robots to merging galaxies
The researchers say the general-purpose framework could be applied to many materials and processes, including soft robot manipulation tasks, RNA evolution and astrophysical phenomena like the merging of galaxies. The research, posted as a preprint on arXiv, will be presented at the ACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia. The researchers want to adapt WorldParticle to simulate these and other physical systems and build on this work toward the development of a full, general-purpose simulation pipeline for scientific discovery.