Research output

Tobias J. Beers — Selected Academic Work

Selected academic and applied research on decision systems, governance and computational methods.

SSRN Working Paper · 2026

Institutional Field Dynamics: A Force Field Model of Multi-Actor Governance

Formalises institutional dynamics as a force field system in which actors move through configuration space under three types of force: intrinsic motivation, social influence mediated by an asymmetric power matrix, and institutional pressure imposed by policy architecture. Policy is modeled as transformation of the field itself. The model generates diagnostic metrics for coherence, friction, and power asymmetry, and enables attractor analysis that tests whether a policy genuinely transforms the institutional landscape or merely shifts positions within it. Demonstrated on Dutch education reform (Passend Onderwijs), with applications to advisory governance, regulatory design, and participatory democracy.

Status Working paper
Year 2026
Institutional dynamics Force field model Multi-actor governance Attractor analysis Policy design
DOI View on SSRN GH Simulation code
Master's Thesis · Computational Science

Using probabilistic methods for hierarchical visualization of single-cell RNA-seq data

Computational methods for making sense of high-dimensional single-cell RNA-seq data. The focus is on probabilistic, hierarchical views that expose structure in cell populations instead of drowning in raw expression matrices.

Institution University of Amsterdam
Year 2020
Single-cell RNA-seq Hierarchical visualization Probabilistic modelling Computational biology
PDF View thesis
Master's Thesis · Information Studies (Data Science)

Using transfer learning to increase performance of multinomial text classification on documents of different municipalities

Transfer learning with TrAdaBoost on heterogeneous municipal document sets. The work looks at how knowledge can be moved between domains to stabilise performance when each city writes in its own bureaucratic dialect.

Institution University of Amsterdam
Year 2018
Transfer learning Text classification TrAdaBoost Naive Bayes
PDF View thesis