arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2302.00032cs.LG
收藏

神经机械自编码器:学习耦合弹性和神经网络非线性

Neuromechanical Autoencoders: Learning to Couple Elastic and Neural Network Nonlinearity

Princeton University(普林斯顿大学)

Deniz Oktay, Mehran Mirramezani, Eder Medina, Ryan P. Adams

摘要:

智能生物系统的特征在于它们在复杂环境中的体现以及它们的神经系统与它们的身体的非线性力学特性之间的密切相互作用。这种协调,其中运动系统的动力学共同进化以减少大脑的计算负担,被称为“机械智能”或“形态计算”。在这项工作中,我们试图开发这个过程的机器学习类似物,我们共同学习复杂的非线性弹性固体的形态以及深度神经网络来控制它。通过使用与传统深度学习架构耦合的弹性力学的专用可微模拟器-我们称之为神经机械自动编码器-我们能够学习通过梯度下降来执行形态学计算。我们的方法的关键是使用机械超材料-细胞固体,特别是-作为形态学基板。正如深度神经网络为感知和控制任务提供了灵活的和非参数化的函数逼近器一样,细胞固体超材料作为一个丰富的和可学习的空间,有希望近似各种驱动任务。在这项工作中,我们利用这些互补的计算概念,共同设计材料和神经网络控制,以实现非直观的机械行为。我们在模拟中演示了如何实现平移,旋转和形状匹配,以及“数字MNIST”任务。我们还制造和评估了其中一个设计,以验证其在现实世界中的行为。

英文摘要:

Intelligent biological systems are characterized by their embodiment in a complex environment and the intimate interplay between their nervous systems and the nonlinear mechanical properties of their bodies. This coordination, in which the dynamics of the motor system co-evolved to reduce the computational burden on the brain, is referred to as ``mechanical intelligence'' or ``morphological computation''. In this work, we seek to develop machine learning analogs of this process, in which we jointly learn the morphology of complex nonlinear elastic solids along with a deep neural network to control it. By using a specialized differentiable simulator of elastic mechanics coupled to conventional deep learning architectures -- which we refer to as neuromechanical autoencoders -- we are able to learn to perform morphological computation via gradient descent. Key to our approach is the use of mechanical metamaterials -- cellular solids, in particular -- as the morphological substrate. Just as deep neural networks provide flexible and massively-parametric function approximators for perceptual and control tasks, cellular solid metamaterials are promising as a rich and learnable space for approximating a variety of actuation tasks. In this work we take advantage of these complementary computational concepts to co-design materials and neural network controls to achieve nonintuitive mechanical behavior. We demonstrate in simulation how it is possible to achieve translation, rotation, and shape matching, as well as a ``digital MNIST'' task. We additionally manufacture and evaluate one of the designs to verify its real-world behavior.

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。

↑