Corpus ID: 231728312

Reservoir Computing with Thin-film Ferromagnetic Devices

  title={Reservoir Computing with Thin-film Ferromagnetic Devices},
  author={Matthew Dale and Richard F. L. Evans and Sarah N. M. Jenkins and Simon E. M. O'Keefe and Angelika Sebald and Susan Stepney and Fernando Torre and Martin A Trefzer},
Advances in artificial intelligence are driven by technologies inspired by the brain, but these technologies are orders of magnitude less powerful and energy efficient than biological systems. Inspired by the nonlinear dynamics of neural networks, new unconventional computing hardware has emerged with the potential for extreme parallelism and ultra-low power consumption. Physical reservoir computing demonstrates this with a variety of unconventional systems from optical-based to spintronic [1… Expand


Reservoir Computing With Spin Waves Excited in a Garnet Film
It is found that the hysteresis characteristics of the spin waves propagating asymmetrically with respect to excitation points works advantageously to realize high diversity in the time-sequential signals in high-dimensional information space, which has the highest significance for effective learning in reservoir computing. Expand
Recurrent neural networks made of magnetic tunnel junctions
Based upon micromagnetic simulation of the magnetization dynamics, it is demonstrated theoretically and numerically that recurrent neural networks consisting of as few as 40 magnetic tunnel junctions can generate and recognize periodic time series after they are trained with an efficient algorithm. Expand
Memristive crossbar arrays for brain-inspired computing
The challenges in the integration and use in computation of large-scale memristive neural networks are discussed, both as accelerators for deep learning and as building blocks for spiking neural networks. Expand
Recent Advances in Physical Reservoir Computing: A Review
An overview of recent advances in physical reservoir computing is provided by classifying them according to the type of the reservoir to expand its practical applications and develop next-generation machine learning systems. Expand
Classification with a disordered dopant-atom network in silicon
The nonlinearity of hopping conduction in a disordered network of boron dopant atoms in silicon is used to perform nonlinear classification and feature extraction, establishing a paradigm of silicon-based electronics for small-footprint and energy-efficient computation. Expand
Reservoir computing using dynamic memristors for temporal information processing
It is shown that the internal ionic dynamic processes of memristors allow the memristor-based reservoir to directly process information in the temporal domain, and it is demonstrated that even a small hardware system with only 88memristors can already be used for tasks, such as handwritten digit recognition. Expand
Neuromorphic computing with nanoscale spintronic oscillators
It is shown experimentally that a nanoscale spintronic oscillator can be used to achieve spoken-digit recognition with an accuracy similar to that of state-of-the-art neural networks and the regime of magnetization dynamics that leads to the greatest performance is determined. Expand
Reservoir Computing Beyond Memory-Nonlinearity Trade-off
A dynamical mechanism behind the memory-nonlinearity trade-off is clarified, which illustrates why nonlinear dynamics degrades memory stored in dynamical system in general and proposes a mixture reservoir endowed with both linear and non linear dynamics that improves the performance of information processing. Expand
Neuromorphic computing with antiferromagnetic spintronics
The prospects and challenges of antiferromagnetic spintronics for neuromorphic computing are discussed and overview and discussion are given on non-spiking artificial neural networks, spiking neural Networks, and reservoir computing. Expand
Vowel recognition with four coupled spin-torque nano-oscillators
The results demonstrate that non-trivial pattern classification tasks can be achieved with small hardware neural networks by endowing them with nonlinear dynamical features such as oscillations and synchronization, and that the high experimental recognition rates stem from the ability of these oscillators to synchronize. Expand