On May 12th, the Salon of Brain Science and Brain-inspired Computing was held at the Xijiao Hotel in the Haidian district, Beijing, jointly hosted by the Institute of Brain Science and Intelligent Technology and the Center for Brain-Inspired Computing Research (CBICR) at Tsinghua University. It was sponsored by the Beijing Science and Technology Information Center, the Electronic Information Field Center of Capital Science and Technology Condition Platform and the Beijing Lingxi Mobile Communication Industry Technology Innovation Alliance. Professor Shi Luping, director of the Centre for Brain-Inspired Computing Research (CBICR) at Tsinghua University was the moderator, and the keynote speakers included Professor Luo Qingming from Huazhong University of Science & Technology, Professor Wang Zhigang from the University of Kent and Professor Huang Tiejun from Peking University. More than 100 experts, scholars and professionals in the industry participated in the salon.

Cao Gang, director of the Beijing Municipal Science & Technology Commission, delivered a wonderful opening speech. On behalf of Beijing Municipal Science & Technology Commission, he emphasized that the salon was crucial for technological development in Beijing and the implementation of national strategy, and that the Commission would support these activities.

Later, the three keynote speakers delivered excellent speeches.

Speech 1: Luo Qingming “Brain Spatial Informatics——A Bridge between Brain Science and Brain-inspired Artificial Intelligence”

Professor Luo Qingming is a doctoral supervisor and the vice president of Huazhong University of Science and Technology (HUST). He was awarded the Distinguished Professor of the Chang Jiang Scholars Program by the Ministry of Education, one of the first awardees in China.

Professor Luo proposed a brand-new concept: Brain Spatial Informatics. This is an integrated science developed to trace, measure, analyze, process and present cross-level, multi-scale brain spatial information. The research on spatial information and features is helpful for us to decode the brain’s functions and brain diseases, and also to accelerate the development of Brain-inspired Artificial Intelligence. He clarified the two key problems facing Visible Brain-wide Networks (VBN)— proper resolution and complete brain imaging. According to Professor Luo, the traditional method takes much more effort and lacks continuity in sections. Information could be easily lost, and it was precise configuration was difficult. He introduced other advanced technologies, namely the vascular brain atlas with a three-dimensional structure, precise imaging techniques for labeling specific neurons through fluorescent staining, and the simultaneous acquisition of neuronal and cellular architecture information. Finally, Professor Luo predicted that the future development of brain spatial informatics would benefit both the research on intracerebral mechanisms and the theoretical support for brain-inspired artificial intelligence.

Speech 2: Wang Zhigang “Building A Brain-inspired Deep Learning Machine”

Professor Wang Zhigang, Head of the School of Computing (formerly known as Computing Laboratory) at the University of Kent, Chairman (UK & Republic of Ireland Chapter) of the IEEE Computer Society and Fellow of the British Computer Society.

Professor Wang briefly reviewed the development of AI and deep learning from Deep Blue to AlphaGo. He stated that the Nobel Prize in Physiology or Medicine in 1981 was a research milestone: since then the core of deep learning, namely the visual cortex of human brain, is of hierarchical structure. Then he stressed the invention and future of Memristors, which was introduced by Leon Chua. Three decades later, it finally became a physical memristor component. According to Wang, the invention of memristor was of great significance to artificial intelligence and deep learning, and was suitable for synaptic function. He pointed out that research into memristors was just at its beginning, and its future application had infinite possibilities: memristor memory, Hebbian Rule, the basis of neural networks, simulating the connection of phenomenon between neurons, adaptive neuromorphic structure realizing the features of memory, prediction and time constant. Thus, a non-Turing and non-programmable method can be used to realize nature-inspired computation; a new HH equation can be proposed by combining the high-order model of the memristor. Professor Wang concluded that Deep Learning would face a lot of challenges in the future.

Speech 3: Huang Tiejun “Neuromorphic Computing”

Professor Huang Tiejun, Head of the Department of Computer Science and Technology at the School of Electronics Engineering and Computer Science at Peking University, a winner of the National Science Fund for Distinguished Young Scholars and distinguished professor of the Chang Jiang Scholars Program of the Ministry of Education.

Professor Huang stated that 2016 is the 10th anniversary of deep learning, the 60th anniversary of Artificial Intelligence, the 70th anniversary of the invention of computers, the 80th anniversary of the Turing Computer Model, and the first year of neuromorphic computing and brain-inspired computers. He first introduced the development of computation in a narrow sense, pointing out that structure was the material basis of function and that the human brain structure justified its superiority over the Turing machine. Professor Huang then reviewed the development of Neuromorphic Computing from the perspective of its research and projects at home and abroad. His current research interest is the general information processing model of the cross-media intelligent cerebral cortex. His research team focuses mainly on the image encoding of neural spiking sequence including simulated retina, video encoding, simulated visual cortex, and image decoding reconfiguration. Using the HTM model for detection, the video of experimental phenomenon shows the obvious change of pulse expression and the high time sensitivity. Finally, Professor Huang shared with the audience his original poems concerning computers, brain science, and brain-inspired research. His concluding remarks connected the three speeches perfectly and were warmly received.

The was a Q & A session right after each keynote speech. The audience enthusiastically engaged with the keynote speakers. These high-level speeches brought a feast of thoughts and ideas and the participants expressed their gratitude for the great benefit they gained at the salon. The salon was full of visions and thoughts, and undoubtedly help promote the development of brain-inspired computing!