Deepan Muthirayan and I have been working on the role of external memory to improve the performance of adaptive controllers in our work at the confluence of machine learning and control. The basic motivation is that memory plays a very big role in learning. However, by and large, when it comes to control, memory is taken to be the state of the (nonlinear) dynamic controller. Our hypothesis is that by explicitly adding external memory to the adaptive control of uncertain, time-varying dynamical systems, the adaptive performance of the closed-loop system can be improved. We have written two papers where we explore this theme. Our results show the promise and potential of this idea. These can be found on the Publications page of this website.
- Memory Augmented Neural Network Adaptive Controllers: Performance and Stability. [Abstract: “In this paper, we propose a novel control architecture, inspired from neuroscience, for adaptive control of continuous-time systems. The proposed architecture, in the setting of standard neural network (NN) based adaptive control, augments an external working memory to the NN. The external working memory, through a write operation, stores certain recently observed feature vectors from the hidden layer of the NN. It retrieves relevant vectors from the working memory to modify the final control signal generated by the controller. The use of external working memory is aimed at improving the context thereby inducing the learning system to search in a particular direction. …”]
- Memory Augmented Neural Network Adaptive Controller for Strict Feedback Nonlinear Systems. [Abstract: … “We propose a novel backstepping memory augmented NN (MANN) adaptive control method for the control of strict feedback non-linear systems. Here, each NN, in the backstepping NN adaptive controller, is augmented with an external working memory. The NN can write relevant information to its working memory and later retrieve them to modify its output, thus providing it with the capability to leverage past learned information effectively and improve its speed of learning. We propose a specific design for this external memory interface and show that the proposed control design achieves bounded stability for the closed loop system. We also provide substantial numerical evidence showing that the proposed memory augmentation improves the speed of learning by a significant margin.”]