Department of Precision Instruments, Center for Brain-Inspired Computing Research (CBICR), Optical Memory National Engineering Research Center, Tsinghua University, Beijing, China.
Beijing Innovation Center for Future Chip, Tsinghua University, Beijing, China.
Nature. 2019 Aug;572(7767):106-111. doi: 10.1038/s41586-019-1424-8. Epub 2019 Jul 31.
There are two general approaches to developing artificial general intelligence (AGI): computer-science-oriented and neuroscience-oriented. Because of the fundamental differences in their formulations and coding schemes, these two approaches rely on distinct and incompatible platforms, retarding the development of AGI. A general platform that could support the prevailing computer-science-based artificial neural networks as well as neuroscience-inspired models and algorithms is highly desirable. Here we present the Tianjic chip, which integrates the two approaches to provide a hybrid, synergistic platform. 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 easily implement brain-inspired circuits and several coding schemes. Using just one chip, we demonstrate the simultaneous processing of versatile algorithms and models in an unmanned bicycle system, realizing real-time object detection, tracking, voice control, obstacle avoidance and balance control. Our study is expected to stimulate AGI development by paving the way to more generalized hardware platforms.
有两种开发通用人工智能 (AGI) 的一般方法：面向计算机科学和面向神经科学。由于它们的公式和编码方案存在根本差异，这两种方法依赖于截然不同且不兼容的平台，从而阻碍了 AGI 的发展。一个能够支持当前基于计算机科学的人工神经网络以及受神经科学启发的模型和算法的通用平台是非常需要的。在这里，我们介绍了“天机芯”芯片，它集成了这两种方法，提供了一个混合的、协同的平台。“天机芯”采用多核架构、可重构模块和混合编码方案的简化数据流，不仅可以容纳基于计算机科学的机器学习算法，还可以轻松实现受大脑启发的电路和几种编码方案。仅使用一块芯片，我们就在无人自行车系统中演示了多种算法和模型的实时处理，实现了实时目标检测、跟踪、语音控制、避障和平衡控制。我们的研究有望通过为更通用的硬件平台铺平道路来刺激 AGI 的发展。
