arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1807.10787cs.LGstat.ML
收藏

基于理论驱动机器学习的近优拓扑一次性生成

One-Shot Generation of Near-Optimal Topology through Theory-Driven Machine Learning

Arizona State University(亚利桑那州立大学);Mechanical Engineering, Arizona State University, Tempe(亚利桑那州立大学机械工程系(坦佩));Mechanical Engineering(机械工程)

Ruijin Cang, Hope Yao, Yi Ren

摘要:

我们引入了一种理论驱动的机制,用于学习一个神经网络模型,该模型在给定问题设置的情况下一次性执行生成拓扑设计,从而规避了计算设计任务通常需要的传统迭代过程。所提出的机制可以导致机器,快速响应新的设计要求的基础上,通过过去的经验积累的设计生成的知识。由于深度神经网络在知识泛化方面的已知限制,通过监督学习实现这样的机制将需要不切实际的大量问题-解决方案对用于训练。为此,我们引入了学生(神经网络)和教师(拓扑优化的最优性条件)之间的交互:学生从现有数据中学习,并在看不见的问题上进行测试。学生的解决方案与最优性条件的偏差被量化,并用于选择新的数据点来学习。我们称这种学习机制为“理论驱动”,因为它明确地使用特定领域的理论来指导学习,从而将其与纯粹的数据驱动的监督学习区分开来。我们表明,通过遵守最小化的问题,建议的学习机制,导致拓扑生成与接近最佳的结构符合性,大大提高了标准的监督学习在相同的计算预算。

英文摘要:

We introduce a theory-driven mechanism for learning a neural network model that performs generative topology design in one shot given a problem setting, circumventing the conventional iterative process that computational design tasks usually entail. The proposed mechanism can lead to machines that quickly response to new design requirements based on its knowledge accumulated through past experiences of design generation. Achieving such a mechanism through supervised learning would require an impractically large amount of problem-solution pairs for training, due to the known limitation of deep neural networks in knowledge generalization. To this end, we introduce an interaction between a student (the neural network) and a teacher (the optimality conditions underlying topology optimization): The student learns from existing data and is tested on unseen problems. Deviation of the student's solutions from the optimality conditions is quantified, and used for choosing new data points to learn from. We call this learning mechanism "theory-driven", as it explicitly uses domain-specific theories to guide the learning, thus distinguishing itself from purely data-driven supervised learning. We show through a compliance minimization problem that the proposed learning mechanism leads to topology generation with near-optimal structural compliance, much improved from standard supervised learning under the same computational budget.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑