Control-Theoretic Techniques for Online Adaptation of Deep Neural Networks in Dynamical Systems

Deep neural networks (DNNs), trained with gradient-based optimization and backpropagation, are currently the primary tool in modern artificial intelligence, machine learning, and data science. In many applications, DNNs are trained offline, through supervised learning or reinforcement learning, and deployed online for inference. However, training DNNs with standard backpropagation and gradient-based optimization gives no intrinsic performance guarantees or bounds on the DNN, which is essential for applications such as controls. Additionally, many offline-training and online-inference problems, such as sim2real transfer of …

applications artificial artificial intelligence backpropagation control cs.lg cs.ne cs.ro cs.sy data data science eess.sy gradient inference intelligence machine machine learning modern networks neural networks offline optimization reinforcement reinforcement learning science standard supervised learning systems through tool training