A recent study conducted by researchers from UCL and Imperial College London has brought us one step closer to achieving brain-inspired computing that consumes less energy. The study, published in the journal Nature Materials, explores the use of chiral (twisted) magnets as a computational medium and demonstrates that by manipulating external magnetic fields and altering temperature, the physical properties of these materials can be adapted to perform various machine-learning tasks.
The approach utilized in the study, known as physical reservoir computing, has been limited in the past due to its lack of reconfigurability. Different materials possess unique physical properties that may excel in certain computing tasks but not in others. The use of chiral magnets, however, proves to be promising as their physical properties can be adjusted to cater to different machine-learning requirements.
Dr. Oscar Lee, the lead author of the paper, explains, “This work brings us a step closer to realizing the full potential of physical reservoirs to create computers that not only require significantly less energy but also adapt their computational properties to perform optimally across various tasks, just like our brains.” The next phase of research involves identifying commercially viable and scalable materials and device architectures.
Traditional computing consumes large amounts of electricity due to the separation of data storage and processing units. This constant shuffling of information results in wasted energy and generates excessive heat. Machine learning, in particular, requires substantial datasets for processing, with training just one large AI model producing hundreds of tonnes of carbon dioxide.
Physical reservoir computing is a neuromorphic approach that aims to eliminate the need for separate memory and processing units, thus enabling more efficient data processing. Aside from its sustainability benefits, physical reservoir computing can be integrated into existing circuitry to provide additional capabilities that are also energy efficient.
The study involved researchers from Japan and Germany, who used a vector network analyzer to assess the energy absorption of chiral magnets under different magnetic field strengths and temperatures ranging from -269 °C to room temperature. The team discovered that different magnetic phases of chiral magnets excelled in various computing tasks. The skyrmion phase, characterized by swirling magnetized particles, demonstrated a strong memory capacity suitable for forecasting tasks. On the other hand, the conical phase exhibited little memory but had exceptional non-linearity, making it perfect for transformation tasks and classification, such as identifying animals.
Co-author Dr. Jack Gartside of Imperial College London highlights the significance of the study, stating, “Our collaborators at UCL recently identified a promising set of materials for powering unconventional computing. These materials can support an especially rich and varied range of magnetic textures. Working with the lead author Dr. Oscar Lee, the Imperial College London group designed a neuromorphic computing architecture to leverage the complex material properties and tailor them to meet the demands of a diverse set of challenging tasks. The results were impressive and demonstrated the direct impact of reconfiguring physical phases on neuromorphic computing performance.”
This research involved the contribution of researchers from the University of Tokyo and Technische Universität München and received support from various institutions including the Leverhulme Trust, Engineering and Physical Sciences Research Council (EPSRC), Imperial College London President’s Excellence Fund for Frontier Research, Royal Academy of Engineering, the Japan Science and Technology Agency, Katsu Research Encouragement Award, Asahi Glass Foundation, and the DFG (German Research Foundation).
*Note:
1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it
Ravina Pandya, Content Writer, has a strong foothold in the market research industry. She specializes in writing well-researched articles from different industries, including food and beverages, information and technology, healthcare, chemical and materials, etc. With an MBA in E-commerce, she has an expertise in SEO-optimized content that resonates with industry professionals.