Thermodynamics of learning physical phenomena
E Cueto, F Chinesta - Archives of Computational Methods in Engineering, 2023 - Springer
Thermodynamics could be seen as an expression of physics at a high epistemic level. As
such, its potential as an inductive bias to help machine learning procedures attain accurate …
such, its potential as an inductive bias to help machine learning procedures attain accurate …
Healthcare As a Service (HAAS): CNN-based cloud computing model for ubiquitous access to lung cancer diagnosis
The field of automated lung cancer diagnosis using Computed Tomography (CT) scans has
been significantly advanced by the precise predictions offered by Convolutional Neural …
been significantly advanced by the precise predictions offered by Convolutional Neural …
Distributed neural network control with dependability guarantees: a compositional port-Hamiltonian approach
Large-scale cyber-physical systems require that control policies are distributed, that is, that
they only rely on local real-time measurements and communication with neighboring agents …
they only rely on local real-time measurements and communication with neighboring agents …
Robust classification using contractive Hamiltonian neural ODEs
Deep neural networks can be fragile and sensitive to small input perturbations that might
cause a significant change in the output. In this letter, we employ contraction theory to …
cause a significant change in the output. In this letter, we employ contraction theory to …
Physically consistent neural ODEs for learning multi-physics systems
Despite the immense success of neural networks in modeling system dynamics from data,
they often remain physics-agnostic black boxes. In the particular case of physical systems …
they often remain physics-agnostic black boxes. In the particular case of physical systems …
Unconstrained parametrization of dissipative and contracting neural ordinary differential equations
D Martinelli, CL Galimberti… - 2023 62nd IEEE …, 2023 - ieeexplore.ieee.org
In this work, we introduce and study a class of Deep Neural Networks (DNNs) in continuous-
time. The proposed architecture stems from the combination of Neural Ordinary Differential …
time. The proposed architecture stems from the combination of Neural Ordinary Differential …
Port-Hamiltonian Neural ODE Networks on Lie Groups For Robot Dynamics Learning and Control
Accurate models of robot dynamics are critical for safe and stable control and generalization
to novel operational conditions. Hand-designed models, however, may be insufficiently …
to novel operational conditions. Hand-designed models, however, may be insufficiently …
LEMURS: Learning distributed multi-robot interactions
This paper presents LEMURS, an algorithm for learning scalable multi-robot control policies
from cooperative task demonstrations. We propose a port-Hamiltonian description of the …
from cooperative task demonstrations. We propose a port-Hamiltonian description of the …
Dynamical systems–based neural networks
Neural networks have gained much interest because of their effectiveness in many
applications. However, their mathematical properties are generally not well understood. If …
applications. However, their mathematical properties are generally not well understood. If …
Universal approximation property of Hamiltonian deep neural networks
M Zakwan, M d'Angelo… - IEEE Control Systems …, 2023 - ieeexplore.ieee.org
This letter investigates the universal approximation capabilities of Hamiltonian Deep Neural
Networks (HDNNs) that arise from the discretization of Hamiltonian Neural Ordinary …
Networks (HDNNs) that arise from the discretization of Hamiltonian Neural Ordinary …