Researchers have developed new neural network architectures for learning complex dynamical systems. One approach focuses on deterministic and stochastic forced Hamiltonian systems, introducing Generalized Forced Hamiltonian Neural Networks (GFHNNs) that offer improved long-time stability and accuracy. Another method addresses piecewise-smooth dynamical systems, which are relevant in fields like climate dynamics and mechanical systems with friction, by combining hyperplane detection with geometry-constrained neural networks. AI
IMPACT These advancements could lead to more accurate and stable models for simulating complex systems in fields like climate science and engineering.
RANK_REASON Two arXiv papers introducing novel neural network architectures for learning dynamical systems.
- arXiv
- Benedikt Brantner
- Climate Dynamics
- dry-friction oscillators
- dynamical systems
- Generalized Forced Hamiltonian Neural Networks
- Lagrange-d'Alembert principle
- machine learning
- Mechanical Systems and Signal Processing
- Parametric Generalized Forced Hamiltonian Neural Networks
- PP04 climate model
- Stratonovich-Taylor expansion
- Wiener process
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
