Track trials, findings, and clinical evidence as it lands.
Search papers, labs, and topics across Lattice.
This paper introduces a probabilistic meta-identification architecture, combining a Transformer encoder-decoder with long-sequence handling, to perform system identification via in-context learning. This approach bypasses explicit model estimation by training a meta model offline on synthetic data to directly map training data and query inputs to outputs. The framework demonstrates competitive accuracy and strong generalization on standard benchmarks and out-of-distribution scenarios, even acting as an uncertainty-aware synthetic data generator for parametric models.
In-context learning now offers calibrated, long-horizon predictions for system identification, rivaling traditional methods and even generating synthetic data for parametric models.
Traditional system identification typically involves two steps: initially, training data is processed to estimate a model of the system of interest; then, the model is used to predict outputs for new input queries. In contrast, recent in-context learning approaches learn a meta model that maps directly from training data and query inputs to outputs in a zero-shot fashion, bypassing explicit model estimation. This approach leverages a meta model trained offline, on large sets of synthetically generated data, to emulate the behavior of a system identification algorithm. In this work, we advance this meta-learning framework for system identification in three directions. First, we develop a unified probabilistic meta-identification architecture that combines a Transformer-based encoder-decoder with tailored mechanisms for handling long training sequences, enabling calibrated predictions over long horizons. Second, we show how a pre-trained meta model can act as an uncertainty-aware synthetic data generator for classical system identification, providing task-specific data that regularize parametric models when real data are scarce. Third, we systematically evaluate the framework on standard nonlinear system identification benchmarks and on a challenging out-of-distribution scenario, demonstrating competitive accuracy with state-of-the-art methods and strong generalization, including cases where input-only adaptation is sufficient to handle significant distribution shifts.