Katz Garrett E, Davis Gregory P, Gentili Rodolphe J, Reggia James A
Department of Electrical Engineering and Computer Science, Syracuse University, Syracuse, NY, United States.
Department of Computer Science, University of Maryland, College Park, MD, United States.
Front Neurorobot. 2021 Dec 14;15:744031. doi: 10.3389/fnbot.2021.744031. eCollection 2021.
We present a neurocomputational controller for robotic manipulation based on the recently developed "neural virtual machine" (NVM). The NVM is a purely neural recurrent architecture that emulates a Turing-complete, purely symbolic virtual machine. We program the NVM with a symbolic algorithm that solves blocks-world restacking problems, and execute it in a robotic simulation environment. Our results show that the NVM-based controller can faithfully replicate the execution traces and performance levels of a traditional non-neural program executing the same restacking procedure. Moreover, after programming the NVM, the neurocomputational encodings of symbolic block stacking knowledge can be fine-tuned to further improve performance, by applying reinforcement learning to the underlying neural architecture.
我们提出了一种基于最近开发的“神经虚拟机”（NVM）的用于机器人操作的神经计算控制器。NVM是一种纯神经循环架构，它模拟了一个图灵完备的、纯符号虚拟机。我们用一种解决积木世界重新堆叠问题的符号算法对NVM进行编程，并在机器人模拟环境中执行它。我们的结果表明，基于NVM的控制器可以忠实地复制执行相同重新堆叠过程的传统非神经程序的执行轨迹和性能水平。此外，在对NVM进行编程之后，通过对底层神经架构应用强化学习，可以对符号积木堆叠知识的神经计算编码进行微调，以进一步提高性能。
