​​​ ​​​ ​

The TWIST research group focuses on two main branches of research.

Topological Spin Textures

We study the physics of topological spin textures such as skyrmions and hopfions, focusing on their theoretical description, dynamical behavior, and realization in real materials. Key themes include current- and field-driven dynamics, emergent electrodynamics, and the interplay of topology with competing interactions at interfaces and in multilayer systems. Our work further addresses how embedding magnetic textures in hybrid material environments can unlock emergent physical phenomena beyond conventional magnetism.

Physics and AI

Our group works at the intersection of artificial intelligence and physics, pursuing two complementary directions: applying machine learning to analyze and optimize complex physical systems ("AI for Physics"), and exploiting intrinsic material dynamics as computational hardware ("Physics for AI"). On the AI for Physics side, we develop interpretable ML methods for magnetism, while on the Physics for AI side we investigate how nonlinear and stochastic material dynamics can serve as energy-efficient systems for neuromorphic and reservoir computing. Our broader goal is to establish physical unconventional computing as a scalable alternative to conventional digital AI hardware.