New SD-DPC framework learns interpretable feedback policies for nonlinear systems

Researchers have developed a new framework called Sparse Dictionary Differentiable Predictive Control (SD-DPC) for learning interpretable feedback policies for nonlinear systems. This method first identifies a prediction model using sparse identification of nonlinear dynamics (SINDy) and then trains a policy as a sparse combination of dictionary functions. The resulting explicit feedback law requires significantly less memory and computation compared to optimization benchmarks, while also demonstrating superior performance and explicit sensitivity bounds across various control problems. AI

IMPACT This framework could lead to more efficient and interpretable control systems in various applications.

RANK_REASON The item is a research paper detailing a new control framework. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.LG →

• Ali Reza Daneshvar Garmroodi
• alphaXiv
• arXiv
• CatalyzeX
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• ScienceCast
• SD-DPC
• SINDy

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →