Skip to content

Current-research flagship

Equifinality & Inverse Generative Social Science

Different generative mechanisms can produce similar observable behavior. This work investigates how machine learning and conformal prediction can be used to identify sets of plausible behavioral rules in agent-based models rather than forcing a single overconfident explanation.

A diagram showing three separate generative mechanisms, each following a different path, converging on a similar observed outcome — illustrating equifinality: the idea that different underlying processes can produce indistinguishable results.

The problem

Agent-based models (ABMs) let researchers specify simple, local behavioral rules and observe the aggregate patterns those rules produce. But this creates an inverse problem: given only the aggregate pattern, many different underlying rule sets can often reproduce it equally well. This property — equifinality — means that fitting a single “best” mechanism to observed data can be misleading, since it hides the range of mechanisms that would have looked just as plausible.

Why equifinality matters

For fields that rely on agent-based modeling to explain social phenomena — from segregation dynamics to information diffusion — treating a fitted rule set as the explanation risks overstating what the data can actually tell us. A more honest scientific stance is to ask: given what we observed, what is the full range of mechanisms that remain plausible?

From observation to mechanism

This project treats mechanism identification as an inference problem: start from simulated or empirical observations, and work backward toward the set of behavioral rules consistent with them, rather than a single point estimate.

Our approach

The approach combines agent-based simulation with machine learning models trained to map observed outcomes back to candidate generative rules, then uses conformal prediction to produce calibrated, uncertainty-aware sets of plausible rules rather than a single prediction.

Conformal prediction and plausible rule sets

Conformal prediction provides a statistically principled way to output a set of plausible answers with a guaranteed coverage rate, rather than a single answer with an unstated (and often overconfident) level of confidence. Applied to rule identification in ABMs, this produces certified sets of behavioral rules that remain consistent with the observed data.

My contribution

Chathura is developing this methodology as lead author, including the framing of rule identification as a conformal-prediction problem, the experimental design for testing it on agent-based segregation models, and the resulting conference presentation and in-preparation manuscripts.

This project extends directly from Chathura’s earlier work on information diffusion and social simulation — both areas where the same underlying question recurs: what process could plausibly have generated what we observe?

My contribution

Chathura is developing the conformal-prediction methodology for identifying plausible rule sets in agent-based models and is lead author on the associated conference presentation and in-preparation manuscript.

Collaborators


Publications

Publications

2026Accepted

Measuring Equifinality: Conformal Rule Identification in Agent-Based Models

Jayalath, C., Rand, W., Garibay, I.

Conference of the Computational Social Science Society of the Americas (CSS), Santa Fe, NM

Introduces a conformal-prediction approach to identifying plausible sets of behavioral rules in agent-based models, rather than a single point estimate.

  • Conformal Prediction
  • Agent-Based Models
  • Equifinality

Verification note: Listed in the CV as accepted and to be presented; confirm final proceedings metadata (page numbers, DOI) once published.

In preparation

Certified Behavioral Rule Discovery in Agent-Based Segregation Models via Conformal Prediction

Jayalath, C., Rand, W., Garibay, I.

In preparation

Applies conformal prediction to certify plausible behavioral rule sets in agent-based models of residential segregation.

  • Conformal Prediction
  • Agent-Based Models
  • Rule Identification

Related

Related research