Neural Physicist:从图像序列中学习物理动力学
Neural Physicist: Learning Physical Dynamics from Image Sequences
摘要:
我们提出了一种名为神经物理学家(NeurPhy)的新架构,使用深度神经网络直接从图像序列中学习物理动力学。对于任何物理系统,给定全局系统参数,状态的时间演化由潜在的物理定律支配。如何以端到端的方式学习有意义的系统表示,并估计准确的状态转换动态,以促进长期预测一直是长期存在的挑战。在本文中,通过利用表示学习和状态空间模型(SSM)的最新进展,我们提出了神经物理,它使用变分自动编码器(VAE)提取潜在的马尔可夫动态状态在每个时间步,神经过程(NP)提取全局系统参数,和一个非线性非递归随机状态空间模型学习的物理动态转换。我们将NeurPhy应用于两个物理实验环境,即,阻尼摆和行星轨道运动,并取得了可喜的成果。我们的模型不仅可以提取物理上有意义的状态表示,而且还可以学习状态转换动态,从而对看不见的图像序列进行长期预测。此外,从潜在状态空间的流形维度,我们可以很容易地识别底层物理系统的自由度(DoF)。
英文摘要:
We present a novel architecture named Neural Physicist (NeurPhy) to learn physical dynamics directly from image sequences using deep neural networks. For any physical system, given the global system parameters, the time evolution of states is governed by the underlying physical laws. How to learn meaningful system representations in an end-to-end way and estimate accurate state transition dynamics facilitating long-term prediction have been long-standing challenges. In this paper, by leveraging recent progresses in representation learning and state space models (SSMs), we propose NeurPhy, which uses variational auto-encoder (VAE) to extract underlying Markovian dynamic state at each time step, neural process (NP) to extract the global system parameters, and a non-linear non-recurrent stochastic state space model to learn the physical dynamic transition. We apply NeurPhy to two physical experimental environments, i.e., damped pendulum and planetary orbits motion, and achieve promising results. Our model can not only extract the physically meaningful state representations, but also learn the state transition dynamics enabling long-term predictions for unseen image sequences. Furthermore, from the manifold dimension of the latent state space, we can easily identify the degree of freedom (DoF) of the underlying physical systems.
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。