Ensemble Kalman filter for uncertainty in human language comprehension

TL;DR

This paper introduces a Bayesian extension of the Ensemble Kalman Filter to improve neural network models of sentence comprehension by better capturing uncertainty, especially in ambiguous or unexpected language inputs.

Contribution

It presents a novel Bayesian inference framework for neural language models, enhancing their ability to model human-like uncertainty during sentence processing.

Findings

Bayesian methods outperform MLE in representing uncertainty.

Enhanced model better mimics human responses to ambiguous sentences.

Numerical experiments validate improved uncertainty quantification.

Abstract

Artificial neural networks (ANNs) are widely used in modeling sentence processing but often exhibit deterministic behavior, contrasting with human sentence comprehension, which manages uncertainty during ambiguous or unexpected inputs. This is exemplified by reversal anomalies-sentences with unexpected role reversals that challenge syntax and semantics-highlighting the limitations of traditional ANN models, such as the Sentence Gestalt (SG) Model. To address these limitations, we propose a Bayesian framework for sentence comprehension, applying an extension of the ensemble Kalman filter (EnKF) for Bayesian inference to quantify uncertainty. By framing language comprehension as a Bayesian inverse problem, this approach enhances the SG model's ability to reflect human sentence processing with respect to the representation of uncertainty. Numerical experiments and comparisons with maximum likelihood estimation (MLE) demonstrate that Bayesian methods improve uncertainty representation, enabling the model to better approximate human cognitive processing when dealing with linguistic ambiguities.