English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We show that a maximum likelihood approach for parameter estimation inagent-based models (ABMs) of opinion dynamics outperforms the typicalsimulation-based approach. Simulation-based approaches simulate the modelrepeatedly in search of a set of parameters that generates data similar enoughto the observed one. In contrast, likelihood-based approaches derive alikelihood function that connects the unknown parameters to the observed datain a statistically principled way. We compare these two approaches on thewell-known bounded-confidence model of opinion dynamics. We do so on threerealistic scenarios of increasing complexity depending on data availability:(i) fully observed opinions and interactions, (ii) partially observedinteractions, (iii) observed interactions with noisy proxies of the opinions.We highlight how identifying observed and latent variables is fundamental forconnecting the model to the data. To realize the likelihood-based approach, wefirst cast the model into a probabilistic generative guise that supports aproper data likelihood. Then, we describe the three scenarios via probabilisticgraphical models and show the nuances that go into translating the model.Finally, we implement the resulting probabilistic models in an automaticdifferentiation framework (PyTorch). This step enables easy and efficientmaximum likelihood estimation via gradient descent. Our experimental resultsshow that the maximum likelihood estimates are up to 4x more accurate andrequire up to 200x less computational time.

Likelihood-Based Methods Improve Parameter Estimation in Opinion Dynamics Models