Tag
This paper introduces a unified decision-theoretic pretraining framework for neural network-based time series estimators, trained on stratified simulations to approximate near-optimal decision rules. Experiments show that the resulting estimators outperform traditional methods like maximum likelihood estimation on both synthetic and real-world benchmarks.
This paper investigates using vision-language models to assess nursing competency from egocentric video during simulation, finding that recognition accuracy inversely relates to competency level, suggesting a pedagogically informative signal.