Summary of Bayesian Calibration Of Stochastic Agent Based Model Via Random Forest, by Connor Robertson et al.
Bayesian calibration of stochastic agent based model via random forest
by Connor Robertson, Cosmin Safta, Nicholson Collier, Jonathan Ozik, Jaideep Ray
First submitted to arxiv on: 27 Jun 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG); Applications (stat.AP)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a random forest-based surrogate modeling technique to accelerate the evaluation of agent-based models (ABMs) in epidemiology, specifically for calibrating an ABM called CityCOVID. The method utilizes dimensionality reduction via temporal decomposition and principal component analysis (PCA), as well as sensitivity analysis. By applying Markov chain Monte Carlo (MCMC) sampling, the authors demonstrate improved predictive performance compared to approximate Bayesian calibration (IMABC). The technique is shown to effectively calibrate CityCOVID’s quantities of interest, including hospitalizations and deaths, by matching COVID-19 numbers in Chicago from March to June 2020. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper uses a special type of computer model called an agent-based model to help understand how diseases like COVID-19 spread. These models are very good at showing how different people interact with each other and their environment, but they can be very slow and hard to use because they need to be calibrated just right. The researchers came up with a new way to make these models run faster by using something called random forests. They tested this method on a model of COVID-19 spreading in Chicago and found that it worked really well. This could help us better understand how diseases spread and how we can stop them. |
Keywords
» Artificial intelligence » Dimensionality reduction » Pca » Principal component analysis » Random forest
