partial differential equation
PulseAugur coverage of partial differential equation — every cluster mentioning partial differential equation across labs, papers, and developer communities, ranked by signal.
- instance of alphaXiv 90%
- instance of Gotit.pub 90%
- instance of ScienceCast 90%
- used by Neural Operators 90%
- developed Neural Operators 90%
- uses Neural Operators 90%
- used by Gotit.pub 70%
- instance of CatalyzeX 70%
- used by IArxiv 70%
- used by CatalyzeX 70%
- developed by Neural Operators 70%
- instance of Neural Operators 70%
9 day(s) with sentiment data
-
TORNADO framework enhances PDE forecasting with adaptive stochasticity
Researchers have introduced TORNADO, a novel framework for probabilistic forecasting of partial differential equations (PDEs). This approach adapts latent stochasticity to local dynamics, improving uncertainty calibrati…
-
New MC^2 solver accelerates PDE solving with neural-Monte Carlo hybrid
Researchers have developed MC$^2$, a novel hybrid solver that combines neural networks with Monte Carlo methods to accelerate the solving of partial differential equations (PDEs). This approach aims to achieve the accur…
-
New Multifidelity Operator Inference method improves ML surrogate models
Researchers have introduced Multifidelity Operator Inference (MF-OpInf), a novel machine learning approach designed to improve surrogate modeling for complex engineering simulations governed by partial differential equa…
-
New sampling method boosts PDE reconstruction accuracy and speed
Researchers have developed a new method called Function Space Backward-Consistent Sampling (FunBCS) to improve the accuracy and speed of reconstructing partial differential equation (PDE) coefficient fields from sparse …
-
MIGONet introduces solver-free training for transport phenomena modeling
Researchers have introduced MIGONet, a novel graph neural operator designed for surrogate modeling of transport phenomena. This model utilizes a unique training approach called discrete manufactured solutions (MMS), whi…
-
New Kriging Method Integrates Global PDE Residuals for Enhanced Probabilistic Solving
Researchers have introduced a novel method called PDE-residual-informed Kriging (PRIK) to enhance the probabilistic framework for solving partial differential equations (PDEs) using Gaussian processes. PRIK directly int…
-
New Velocity Matching method bypasses simulation for learning dynamics
Researchers have introduced Velocity Matching (VM), a novel simulation-free objective for training continuous-time dynamics models. This method aims to overcome limitations of traditional trajectory-based objectives, su…
-
Physics-Guided Flow Matching enhances AI for Total Electron Content forecasting
Researchers have developed a novel generative AI framework called Physics-Guided Flow Matching to improve the accuracy of forecasting global total electron content (TEC). This method integrates physical dynamics, descri…
-
New bounds derived for Active Simple Hypothesis Testing
This paper introduces new upper and lower bounds for the optimal error exponent in Active Simple Hypothesis Testing (ASHT). The research combines a recent Partial Differential Equation (PDE) characterization of the opti…
-
New Physics-Informed Neural Network Tackles High-Dimensional Detection Problems
Researchers have developed QD-PINN, a novel framework utilizing Physics-Informed Neural Networks (PINNs) to address the challenges of Bayesian Quickest Detection problems in high-dimensional diffusion processes. This ap…
-
New GAV environment enhances coding agents for partial differential equations · 1 source tracked
Researchers have developed Generator-Agent-Verifier (GAV), a novel scientific experimentation environment designed to improve the reliability of coding agents working with partial differential equations (PDEs). GAV cons…
-
New regularization technique improves neural PDE solver consistency
Researchers have introduced Align Latent Rollout (ALR), a novel regularization technique designed to enhance the consistency of neural partial differential equation (PDE) solvers. This method addresses the issue of appr…
-
New SALTO framework enhances nonlinear event inference using Koopman theory
Researchers have developed SALTO, a novel framework based on Koopman theory for inferring finite-amplitude sources in nonlinear systems. This method improves upon traditional adjoint techniques by enabling the predictio…
-
New AI architecture improves PDE surrogate models using mixture of experts
Researchers have developed a novel architecture for steady partial differential equation (PDE) surrogate models on unstructured meshes, utilizing a mixture of experts approach. This method partitions the solution based …
-
New method enhances neural operators with learnable derivative orders
Researchers have developed a novel approach called "Fractional is Better: Learnable Derivative Orders in Neural Operator Learning" that enhances neural operator learning for solving partial differential equations. This …
-
New SBNO Framework Solves PDEs with Limited Data
Researchers have introduced SBNO, a novel framework designed for solving partial differential equation (PDE) problems, particularly when information is incomplete. This equation-free approach utilizes diffusion Schrödin…
-
New F$^3$NO neural operator enhances PDE forecasting accuracy
Researchers have developed a new neural operator model called F$^3$NO, designed to improve the accuracy and resolution of partial differential equation (PDE) forecasting. This model decomposes frequency information, all…
-
LLMs adapt neural operator fine-tuning using prior knowledge and feedback
A new study published on arXiv investigates how Large Language Models (LLMs) adapt their decision-making processes when fine-tuning neural operators. Researchers found that LLMs utilize both an initial task-dependent pr…
-
New SED-MCTS method enhances interpretability in PDE solution discovery
Researchers have developed SED-MCTS, a novel Monte Carlo tree search method designed to improve the interpretability and efficiency of discovering symbolic expressions for physical fields from observational data. Unlike…
-
New ORACLE method improves constrained learning by assessing post-optimizer updates
A new research paper introduces ORACLE, a novel method for constrained learning that assesses constraint compatibility after the optimizer has acted on the parameters. This approach, termed optimizer-relative constraine…