logo
ResearchBunny Logo
Learning stochastic dynamics and predicting emergent behavior using transformers

Physics

Learning stochastic dynamics and predicting emergent behavior using transformers

C. Casert, I. Tamblyn, et al.

This innovative research by Corneel Casert, Isaac Tamblyn, and Stephen Whitelam reveals how a transformer neural network, originally created for processing language, can master the complex dynamics of stochastic systems by simply observing a trajectory. Their groundbreaking work predicts unseen emergent behaviors, opening doors to understanding complex systems without traditional modeling techniques.

00:00
Playback language: English
Abstract
This paper demonstrates that a transformer neural network, initially designed for natural language processing, can effectively learn the dynamic rules of a stochastic system by observing a single trajectory. The network accurately predicts emergent behavior under conditions unseen during training. Using a lattice model of active matter, the transformer learns the dynamics without explicit rate enumeration or configuration space coarse-graining, making it applicable to complex systems.
Publisher
Nature Communications
Published On
Feb 29, 2024
Authors
Corneel Casert, Isaac Tamblyn, Stephen Whitelam
Tags
transformer neural network
stochastic systems
emergent behavior
active matter
machine learning

Related Publications

Explore these studies to deepen your understanding of the subject.

Predicting 1, 2 and 3 year emergent referable diabetic retinopathy and maculopathy using deep learning
Medicine and Health

Predicting 1, 2 and 3 year emergent referable diabetic retinopathy and maculopathy using deep learning

P. Nderitu, J. M. N. D. Rio, et al.

Nondestructive monitoring of annealing and chemical-mechanical planarization behavior using ellipsometry and deep learning
Engineering and Technology

Nondestructive monitoring of annealing and chemical-mechanical planarization behavior using ellipsometry and deep learning

Q. Sun, D. Yang, et al.

Synchronously Predicting Tea Polyphenol and Epigallocatechin Gallate in Tea Leaves Using Fourier Transform-Near-Infrared Spectroscopy and Machine Learning
Food Science and Technology

Synchronously Predicting Tea Polyphenol and Epigallocatechin Gallate in Tea Leaves Using Fourier Transform-Near-Infrared Spectroscopy and Machine Learning

S. Ye, H. Weng, et al.

Design of target specific peptide inhibitors using generative deep learning and molecular dynamics simulations
Chemistry

Design of target specific peptide inhibitors using generative deep learning and molecular dynamics simulations

S. Chen, T. Lin, et al.

Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny