Artificial intelligence (AI) systems are getting more powerful, but they consume a huge amount of energy. In contrast, the human brain is small, but remarkably efficient and smart. To equip machines with similar characteristics, scientists are now exploring bioinspired computing (BIC), so that machines can operate fast and energy-efficiently just like how brain do.
A comprehensive review article recently published by a large team of 73 researchers from 49 universities and research institutions spanning Asia, Europe, and North America lays out a detailed roadmap for the BIC hardware vision. The article titled “Technology Roadmap of Bioinspired Computing Hardware” has been published in ACS Nano. Prof. CHAI Yang, Director of Research Institute for Artificial Intelligence of Things (RIAIoT), Management Committee Member of Research Institute for Intelligent Wearable Systems (RI-IWEAR), Member of Photonics Research Institute (PRI) and Associate Dean of the Faculty of Science, is the corresponding author. His postdoctoral fellow, Dr. WANG Shuang, is the first author.
The article highlights how BIC offers a promising alternative by emulating the intrinsic advantages of biological systems, such as parallelism, adaptability and robustness. Progress in BIC hardware requires interdisciplinary convergence, bridging materials science and device physics with neuroscience, computer science, mathematics and information science. Consequently, the development of this interdisciplinary field urgently requires a comprehensive roadmap that systematically and thoroughly analyses frontier issues and the latest progress.
The roadmap categorises the critical challenges into three components, namely, hardware foundations, architectures and prototype realisations. It highlights how biological features inspire the design of BIC hardware through device physics and discusses performance metrics and engineering challenges. The article describes how diverse signalling rules and structural organisations in BIC architectures support specific computational prototypes, including electronic and photonic BIC chips, and present a technological roadmap outlining opportunities to expand the functional scope of BIC hardware through coordinated advances in devices, architectures and system demonstrations. This ongoing convergence of interdisciplinary knowledge can help accelerate the transition towards high-efficiency AI hardware.
The review article marks a significant milestone in the field of BIC, setting out a clear vision for future research, and identifying both challenges and opportunities that will shape the next generation of AI hardware.
Read the full review article: https://pubs.acs.org/doi/full/10.1021/acsnano.5c17087
You may also like
