Recurrent network classifier for ultrafast skyrmion dynamics
arXiv:1907.01814 · doi:10.1103/PhysRevApplied.12.054026
Abstract
By using the supervised learning we train a recurrent neural network to recognize and classify ultrafast magnetization processes realized in two-dimensional nanosystems with Dzyaloshinskii-Moriya interaction. Our focus is on the different types of skyrmion dynamics driven by ultrafast magnetic pulses. Each process is represented as a sequence of the sorted magnetization vectors inputted into the network. The trained network can perform an accurate classification of the skyrmionic processes at zero temperature in wide ranges of parameters that are the magnetic pulse width and damping factor. The network performance is also demonstrated on different types of unseen data including finite temperature processes. Our approach can be easily adapted for creating an autonomous control system on skyrmion dynamics for experiments or device operations.
arXiv comment7 pages, 11 figures
References in corpus (11)
- Time-resolved Observation and Control of Superexchange Interactions with Ultracold Atoms in Optical Lattices
- A method for atomistic spin dynamics simulations: implementation and examples
- Spin-flip processes and ultrafast magnetization dynamics in Co - unifying the microscopic and macroscopic view of femtosecond magnetism
- Memory functions of magnetic skyrmions
- Femtosecond dynamics of the collinear-to-spiral antiferromagnetic phase transition in CuO
- Nanoscale mapping of ultrafast magnetization dynamics with femtosecond Lorentz microscopy
- Learning Disordered Topological Phases by Statistical Recovery of Symmetry
- Supervised learning magnetic skyrmion phases
- Duplication, collapse and escape of magnetic skyrmions revealed using a systematic saddle point search method
- Learning magnetization dynamics
- Profile approach for recognition of three-dimensional magnetic structures