To create a global platform to unify different approaches in the development of artificial general intelligence (AGI), Chinese researchers, led by Tsinghua University professor Shi Luping, have designed a new type of computer chip called Tianjic. 

The research and development were conducted by Shi Luping and his colleagues at the Center for Brain-Inspired Computing Research of Tsinghua University and was published in the Nature Magazine.

One of the major roadblocks in developing artificial general intelligence (AGI) is the existence of two different approaches to developing it. They are computer science-oriented and neuroscience oriented approaches, with both being fundamentally different in their formulations and coding schemes. The latest Tianjic chip integrates the two methods into one hybrid platform, which enables it to accommodate both machine-learning algorithms and as well as brain-inspired circuits.

The Tianjic chip adopts a many-core architecture, reconfigurable building blocks and a streamlined dataflow with hybrid coding schemes, and can not only accommodate computer-science-based machine-learning algorithms but also quickly implement brain-inspired circuits and several coding schemes, writes the researchers. 

“The Tianjic chip aims to provide a more efficient, fast and flexible computing platform for the research of artificial general intelligence, as well as the development of various applications,” said Shi said Luping to the China Daily.