Innovation
Methods, ideas, or system contributions that make the work stand out.
Neural Event-Mechanism Simulator
Executable Event-Mechanism Priors
Control-Conditioned Multi-Event Dynamics
Kinetic Monte Carlo Benchmark
Physics-Guided Surrogate
NEMSim: Learning Control-Conditioned Multi-Event Physical Dynamics via Executable Event-Mechanism Priors
🤖 AI Summary
This study addresses the computational expense of high-fidelity multi-event physics simulations and the poor generalizability of data-driven approaches by proposing NEMSim, a novel neural simulator. NEMSim pioneers the compilation of predefined event attribute descriptions into executable transition structures, integrating control dependency strength with mechanistic attribution to enable dynamic prediction. This approach transcends the limitations of both traditional equation-constrained and purely data-driven paradigms. Leveraging techniques such as neural networks, kinetic Monte Carlo (KMC) benchmark construction, and event rule compilation, experimental results demonstrate that NEMSim reduces RMSE by 58.9%–81.3%. Furthermore, it achieves state-of-the-art performance in small-sample scenarios, validating the significant gains afforded by its rule-integration mechanism.
📝 Abstract
High-fidelity simulation of control-conditioned multi-event physical systems is computationally expensive, especially across broad control spaces and long trajectories. In these systems, macroscopic evolution emerges from localized discrete events whose intensities and effects depend on process controls and evolving local states, while the available system knowledge is typically expressed as event-attribute descriptions. Purely data-driven surrogates must infer these event effects from limited trajectory coverage, which can hinder generalization to unseen control regimes. Physics-guided methods instead primarily build on equation-level constraints or differentiable solvers rather than discrete event-rule priors. We therefore propose NEMSim (Neural Event-Mechanism Simulator), which compiles predefined event-attribute descriptions into an executable transition structure linking control-dependent event intensities, prior-guided mechanism attribution, and state-dependent responses. To enable evaluation of control-conditioned multi-event dynamics with explicit system knowledge, we construct a 3D KMC-based benchmark pairing high-fidelity trajectories with explicit event rules, standardized splits, and evaluation protocols. Across three settings, NEMSim reduces Avg. RMSE by 58.9%-81.3% relative to the strongest baseline in each setting. It also remains best in the data-efficiency study with training-data fractions down to 10%. Mechanism analyses further show that these gains arise from executable rule integration rather than prior access or architecture alone.
🔎 Similar Papers
2024-07-16Citations: 0
