We present a method for unsupervised learning of equations of motion for objects in raw and optionally distorted unlabeled synthetic video (or, more generally, for discovering and modeling predictable features in time-series data). We first train an autoencoder that maps each video frame into a low-dimensional latent space where the laws of motion are as simple as possible, by minimizing a combination of nonlinearity, acceleration, and prediction error. Differential equations describing the motion are then discovered using Pareto-optimal symbolic regression. We find that our pre-regression ("pregression") step is able to rediscover Cartesian coordinates of unlabeled moving objects even when the video is distorted by a generalized lens. Using intuition from multidimensional knot theory, we find that the pregression step is facilitated by first adding extra latent space dimensions to avoid topological problems during training and then removing these extra dimensions via principal component analysis. An inertial frame is autodiscovered by minimizing the combined equation complexity for multiple experiments.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1103/PhysRevE.103.043307 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Copper(II) complexes of 2-hydroxy-1-naphthaldehyde Schiff bases: synthesis, activity and computational studies.
Future Med Chem
January 2025
Department of Chemistry, University of Malakand, Dir Lower, Khyber Pakhtunkhwa, Pakistan.
Background: Due to the divers biological applications of Cu(II) complexes, we in this study reports the various Cu(II) complexes. The study aims to synthesize and assess new Cu(II) complexes as powerful β-glucuronidase inhibitors.
Methods: Five Schiff base ligands and their complexes were synthesized, characterized, and screened against β-glucuronidase inhibitory activity.
A guidance to intelligent metamaterials and metamaterials intelligence.
Nat Commun
January 2025
ZJU-UIUC Institute, Interdisciplinary Center for Quantum Information, State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang University, Hangzhou, China.
The bidirectional interactions between metamaterials and artificial intelligence have recently attracted immense interest to motivate scientists to revisit respective communities, giving rise to the proliferation of intelligent metamaterials and metamaterials intelligence. Owning to the strong nonlinear fitting and generalization ability, artificial intelligence is poised to serve as a materials-savvy surrogate electromagnetic simulator and a high-speed computing nucleus that drives numerous self-driving metamaterial applications, such as invisibility cloak, imaging, detection, and wireless communication. In turn, metamaterials create a versatile electromagnetic manipulator for wave-based analogue computing to be complementary with conventional electronic computing.
View Article and Find Full Text PDFRadiogenomic explainable AI with neural ordinary differential equation for identifying post-SRS brain metastasis radionecrosis.
Med Phys
January 2025
Deparment of Radiation Oncology, Duke University, Durham, North Carolina, USA.
Background: Stereotactic radiosurgery (SRS) is widely used for managing brain metastases (BMs), but an adverse effect, radionecrosis, complicates post-SRS management. Differentiating radionecrosis from tumor recurrence non-invasively remains a major clinical challenge, as conventional imaging techniques often necessitate surgical biopsy for accurate diagnosis. Machine learning and deep learning models have shown potential in distinguishing radionecrosis from tumor recurrence.
View Article and Find Full Text PDFLensless imaging offers a lightweight, compact alternative to traditional lens-based systems, ideal for exploration in space-constrained environments. However, the absence of a focusing lens and limited lighting in such environments often results in low-light conditions, where the measurements suffer from complex noise interference due to insufficient capture of photons. This study presents a robust reconstruction method for high-quality imaging in low-light scenarios, employing two complementary perspectives: model-driven and data-driven.
View Article and Find Full Text PDFThree-dimensional (3D) light-field displays can provide natural stereoscopic visual perception and an intuitive viewing experience. However, the high production threshold and the lack of user-friendly editing tools for light-field images make it difficult to efficiently and conveniently generate 3D light-field content that meets various needs. Here, a text-driven light-field content editing method for 3D light-field display based on Gaussian splatting is presented.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!