Myriad: a real-world testbed to bridge trajectory optimization and deep learning
This work addresses the need for bridging trajectory optimization and deep learning in real-world continuous environments, offering a tool for the machine learning community to apply modern techniques to impactful tasks, though it is incremental as it builds on existing methods.
The authors introduced Myriad, a JAX-based testbed that provides trajectory optimization techniques and real-world optimal control problems for machine learning practitioners, and demonstrated a novel end-to-end model using implicit planning over neural ODEs for learning and control tasks.
We present Myriad, a testbed written in JAX for learning and planning in real-world continuous environments. The primary contributions of Myriad are threefold. First, Myriad provides machine learning practitioners access to trajectory optimization techniques for application within a typical automatic differentiation workflow. Second, Myriad presents many real-world optimal control problems, ranging from biology to medicine to engineering, for use by the machine learning community. Formulated in continuous space and time, these environments retain some of the complexity of real-world systems often abstracted away by standard benchmarks. As such, Myriad strives to serve as a stepping stone towards application of modern machine learning techniques for impactful real-world tasks. Finally, we use the Myriad repository to showcase a novel approach for learning and control tasks. Trained in a fully end-to-end fashion, our model leverages an implicit planning module over neural ordinary differential equations, enabling simultaneous learning and planning with complex environment dynamics.