Causal models improve decision making and system dynamics understanding
Towards Realistic Models of Change: A Causal Perspective on Decision-Making and Dynamical Systems
Machine LearningArtificial Intelligence
Summary
Statistical methods often struggle when conditions change, because they rely only on data they have seen before. The authors study how using causal models, which focus on understanding what causes what, can help systems make better decisions and predict how things change over time. They look at helping people figure out how to change things to get better decisions and also at how systems that evolve over time can be understood with causal ideas. Their goal is to build models that work well in the real world by realistically representing how changes actually happen.
What this means in practice
- •For risk assessment teams: Design decision-support tools that suggest actionable and realistic changes individuals can make to influence unfavorable outcomes, considering timing and true impact.
- •For control system engineers: Model and analyze how interventions affect the equilibrium states of complex systems that evolve over time with uncertainty, to improve robustness under changing conditions.
A position paper. It proposes an approach and reports no results.
Authors
Isacco Beretta
Abstract
While statistical learning has made remarkable progress in recent decades, driven by large datasets and data-driven protocols that have transformed scientific practice, it still faces a key limitation: it struggles to generalize beyond the data it has seen, often failing under distribution shift. Causality offers a promising solution: causal models focus on the underlying mechanisms of change rather than surface-level associations, enabling robust reasoning under external interventions. This allows systems not only to learn from data, but also to reason about hypothetical scenarios, adapt to new environments, and inform better decisions. This thesis explores how these approaches can inform each other, in two parts. The first part examines algorithmic recourse, that is, identifying which changes an individual can make to overturn an unfavorable decision. We propose an alternative perspective to the established framework, arguing for a more realistic cost function, a reformulation of the problem in causal terms, and the inclusion of a temporal dimension in policy evaluation. The second part focuses on dynamical systems, extending the notions of causal interventions and effects to discrete and continuous stochastic processes. We show that the equilibria of these processes, and how they change following an intervention, can be linked to Structural Causal Models (SCMs), viewed as compressed representations of the underlying temporal dynamics. We argue for the relevance of such temporal causal models, since in real-world scenarios actions take time to produce their effects. The common goal of both parts is practical: the utility, robustness, and trustworthiness of these models depend critically on their ability to realistically represent the underlying phenomena.