Get Started
Home
Topics
Search
Library
Research questionHow can process simulators preserve latent-state uncertainty when generating trajectories from partial event logs?Event logs may omit service-start times and other state information, so one observed process prefix can correspond to several underlying conditions. A simulator that represents only one recurrent state can miss plausible routing and timing behavior.
AI
Health
Machine Learning
Research Paper
Statistical Machine Learning
Latest papersRecent research connected to this question, newest first.A Unified Particle Filter LSTM for Data-Driven Process SimulationThe evidence concerns a Unified Particle Filter LSTM trained end-to-end on event logs and evaluated on three real-world emergency-department datasets. Reported outcomes cover next-activity prediction, conditional sojourn-time quantiles, and system-level behavior; the abstract does not establish performance beyond those datasets.research paper · Sep 2, 2026
Related questions
How can learned graph simulators preserve contact history for accurate long-horizon granular-dynamics rollouts?How can we estimate simulator-parameter distributions that reproduce observations without inaccurate fixed likelihood surrogates?How can compact device surrogates stay physically consistent and accurate across unseen process or geometry splits with scarce data?How can low-precision recurrent-state storage preserve small updates during temporal inference?