New framework CIPHER learns latent dynamics in Port-Hamiltonian systems

Researchers have developed CIPHER, a novel framework for learning latent dynamics in Port-Hamiltonian systems from partial observations. This two-stage approach uses contrastive learning to first identify predictive state representations and then learns the system's dynamics. CIPHER demonstrates competitive or superior performance compared to existing methods on various datasets, showing robustness and improved long-horizon forecasting for complex systems. AI

IMPACT This research advances methods for learning complex system dynamics from partial data, potentially improving forecasting and control in scientific and engineering applications.

RANK_REASON The cluster contains a research paper detailing a new method for learning system dynamics. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• CIPHER
• DagsHub
• Gotit.pub
• Hugging Face
• IArxiv
• Neural ODE
• Peilun Li
• ScienceCast

• paper
• model release

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