Burgers
PulseAugur coverage of Burgers — every cluster mentioning Burgers across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
Function-Space Transformer Adapts Representation for Scientific and Visual Tasks
Researchers have introduced the Function-Space Transformer (FST), a novel framework designed to learn from functions represented by discrete samples. Unlike traditional methods that use fixed grids, FST employs a spatia…
-
New PB--SAV optimizer enhances scientific machine learning objectives
Researchers have developed a new optimization method called the pullback-corrected scalar auxiliary variable (PB--SAV) optimizer, designed for complex objectives in scientific machine learning. This method uses a scalar…
-
New protocol tackles data heterogeneity in federated learning for PDEs
A new protocol called PDE-Dirichlet has been developed to address data heterogeneity in federated learning for scientific machine learning tasks involving partial differential equations (PDEs). This protocol quantifies …
-
New SS-ESOAP method enhances physics-informed neural network training
Researchers have developed SS-ESOAP, a novel preconditioning method designed to improve the training of physics-informed neural networks (PINNs). This method addresses the challenge of ill-conditioned objectives that of…
-
New MPNO model enhances stability in transient dynamics prediction
Researchers have developed a new neural operator called the Constitutive Markov Physics-Informed Neural Operator (MPNO) designed to improve stability in predicting transient dynamics, particularly for problems with stro…
-
New method enhances equation discovery from noisy data using Koopman dynamics
Researchers have developed a dynamics-aware method for identifying governing equations from sparse and noisy data, building upon techniques like Sparse Identification of Nonlinear Dynamics (SINDy) and PDE Functional Ide…
-
Image editing models show potential as unified numerical solvers
Researchers have explored the potential of using pretrained generative image-editing models as a unified interface for numerical simulations. By rendering physical inputs and solutions as images and using lightweight ad…
-
New TGSR-PINN method enhances physics-informed neural network transfer learning
Researchers have developed a new method called Target-Guided Selective Reweighting PINN (TGSR-PINN) to improve the transfer learning capabilities of physics-informed neural networks (PINNs) for inverse problems. This ap…
-
New neural network methods tackle complex partial differential equations · 3 sources tracked
Researchers have developed new neural network frameworks for solving partial differential equations (PDEs) in complex domains. One approach, Domain-Decomposed Randomized Neural Networks, uses specialized subnetworks for…
-
AI designs sustainable and nutritious burgers
Researchers have utilized generative artificial intelligence to design novel burger recipes that are both delicious and sustainable. The AI model analyzed various ingredients and their nutritional profiles to create opt…
-
New HSPINN method enhances physics-informed neural network accuracy
Researchers have developed a new method called Adaptive Hard-Soft Physics-Informed Neural Networks (HSPINN) to improve the training and accuracy of physics-informed neural networks (PINNs). Traditional PINNs struggle wi…
-
New testing method validates scientific ML surrogates
Researchers have developed a new method for testing scientific machine-learning (SciML) surrogates, which approximate complex simulations. The proposed approach, called Domain-Validity-Gated Metamorphic Testing, address…
-
New PDE framework offers stable, efficient solutions without traditional methods
Researchers have developed a novel framework for solving partial differential equations (PDEs) that bypasses traditional matrix-based methods and data-intensive neural network training. This new approach utilizes physic…