Ordinary Differential Equations
PulseAugur coverage of Ordinary Differential Equations — every cluster mentioning Ordinary Differential Equations across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New model identifies parameters of human learning dynamics
Researchers have developed a new model to understand human learning as a dissipative dynamical process, akin to a nonlinear system of ordinary differential equations. This model aims to capture how knowledge accumulates…
-
New ML method learns unknown ODEs using functional analysis
Researchers have developed a novel machine learning method for discovering unknown nonlinear ordinary differential equations (ODEs) from a single state trajectory. This approach is grounded in functional analysis and op…
-
LLM framework LLMODE enhances irregular spatio-temporal forecasting
Researchers have developed LLMODE, a novel framework designed to improve spatio-temporal forecasting using large language models (LLMs). This method addresses limitations in existing approaches, such as handling irregul…
-
New framework models nonlinear delay dynamics using Koopman operator
Researchers have developed a novel framework for approximating the Koopman operator of nonlinear delay differential equations (DDEs). This approach bridges the gap between infinite-dimensional DDE dynamics and finite-di…
-
Autonomous driving planner uses flow matching for real-time control
Researchers have developed a new flow-matching planner for autonomous driving that directly generates control trajectories, including acceleration and curvature profiles. This model is conditioned on a bird's-eye-view r…
-
New method stabilizes drone motion using integro-differential equations
Researchers have developed a new method for stabilizing drone motion using distributed feedback control, which involves an integral operator with potentially unbounded memory. This approach allows for the study of integ…
-
Liquid Neural Networks offer low-compute alternative to LLMs
Liquid Neural Networks (LNNs) offer an alternative to Large Language Models (LLMs) by utilizing continuous-time dynamics based on Ordinary Differential Equations (ODEs) rather than discrete symbol processing. Unlike LLM…
-
New hybrid framework learns physics ODEs with neural networks
Researchers have developed a novel hybrid framework that combines neural networks with physics-based ordinary differential equations (ODEs) to model dynamical systems. This method is particularly useful when some ODE co…
-
LLM-powered agent discovers biological ODEs with symbolic regression
Researchers have developed MEDA, a new system that combines large language models (LLMs) with symbolic regression to automatically discover Ordinary Differential Equations (ODEs) for biological systems. This agentic fra…
-
New neural network training scheme uses gradient flows and Lojasiewicz theory
Researchers have developed a new training scheme for neural networks that utilizes analytic activation functions and is based on gradient flows. This method, which guarantees convergence through Lojasiewicz theory, offe…
-
PINN algorithm enhances parachute line deployment analysis · 2 sources tracked
Researchers have developed a physics-informed neural network (PINN) algorithm to predict tension during parachute suspension line deployment. This method offers improved computational efficiency and accuracy compared to…
-
New flow models for graph signals offer enhanced stability
Researchers have analyzed continuous normalized flow models for graph signal generation, demonstrating that permutation equivariance is maintained in both continuous-time ordinary differential equations and their discre…
-
LLM-guided framework discovers ODEs from aggregate data
Researchers have developed AgentODE, a novel framework designed to discover ordinary differential equation (ODE) structures and infer parameter distributions from aggregate data, particularly for rare diseases where ind…
-
Differential Equations Inspire New Deep Neural Network Architectures
A new paper explores the integration of differential equations with deep neural networks to enhance theoretical understanding, interpretability, and generalization capabilities in AI. The research reviews architectures …
-
New theory bounds ODE identification from solution data
Researchers have developed a new theoretical framework for identifying governing equations from solution data, addressing a fundamental challenge in scientific machine learning. The approach introduces the Hausdorff dis…
-
LLM-ACES framework uses large language models to discover dynamical systems
Researchers have developed LLM-ACES, a novel framework that uses large language models to guide the discovery of dynamical systems by searching for Ordinary Differential Equations (ODEs). This closed-loop system optimiz…
-
New SINDy Method Discovers Dynamical Systems from Noisy, Multi-Fidelity Data
Researchers have developed a new method called Multi-Fidelity SINDy to discover nonlinear dynamical systems from data with varying levels of noise and fidelity. This approach extends the existing Sparse Identification o…
-
New bounds improve error estimation for physics-informed neural networks
Researchers have developed new methods for estimating errors in Physics-Informed Neural Networks (PINNs), which are used to solve differential equations by combining machine learning with physical laws. The work introdu…
-
Sakana AI's DiffusionBlocks cuts training memory by training network blocks independently
Sakana AI has introduced DiffusionBlocks, a novel framework for training neural networks more efficiently. This method partitions a network into multiple blocks, allowing each block to be trained independently. By reduc…
-
New solver tackles ODEs with single-trajectory signals
Researchers have developed a novel branched signature kernel solver designed to accurately model ordinary differential equations (ODEs) driven by single, potentially rough, trajectory signals. This new method addresses …