Brain-inspired architectures for next-generation microelectronic systems

下一代微电子系统的类脑架构

• 批准号: EP/G035806/1
• 负责人: Piotr Dudek
• 金额: $ 4.26万
• 依托单位: University of Manchester
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2009
• 资助国家: 英国
• 起止时间: 2009 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FG035806%2F1
• 关键词: Brain; inspired; architectures; next; generation

Brain-like information processing systems, based on massive parallelism, non-linear dynamics and complex interactions between simple cells, may provide novel computer architectures suitable for engineering electronic systems of the future. The aim of this research visit to Hong Kong University of Science and Technology is to work, in collaboration with the group of Prof. Bertam Shi, on assessing the feasibility of a number of approaches to brain-inspired computation with respect to their implementability in present-day and emerging integrated circuit technologies. Based on the shared experiences and expertise in this area, through extensive study of existing models, literature reviews, requirements analysis and exploratory designs, we will identify the most promising approaches that will be taken further in future research projects.

类脑信息处理系统基于大规模并行性、非线性动力学和简单细胞之间的复杂相互作用，可以提供适合于未来工程电子系统的新型计算机架构。是次访问香港科技大学的目的，是与施伯坦教授的研究小组合作，评估多项脑启发计算方法在现今及新兴集成电路技术中的可行性。基于在这一领域的经验和专业知识，通过对现有模型的广泛研究，文献综述，需求分析和探索性设计，我们将确定最有前途的方法，将在未来的研究项目中进一步采取。

项目成果

Self-Organizing Neural Population Coding for improving robotic visuomotor coordination

用于改善机器人视觉运动协调的自组织神经群体编码

• DOI: 10.1109/ijcnn.2011.6033393
• 发表时间: 2011
• 作者: Zhou T

Artificial Neural Networks - ICANN 2010

人工神经网络 - ICANN 2010

• DOI: 10.1007/978-3-642-15822-3_23
• 发表时间: 2010
• 作者: Brohan K

Binocular disparity calculation on a massively-parallel analog vision processor

大规模并行模拟视觉处理器上的双目视差计算

• DOI: 10.1109/cnna.2010.5430282
• 发表时间: 2010
• 作者: Mandai S

Development of robot self-identification based on visuomotor prediction

基于视觉运动预测的机器人自我识别开发

• DOI: 10.1109/devlrn.2012.6400878
• 发表时间: 2012
• 作者: Tao Zhou