Abstract In this paper we compare causal models with reason in the construction of Bayesian networks for legal evidence. models, arrows network are drawn from causes to effects. a model, instead towards evidence, factum probandum probans . We explore differences between and observe several distinct advantages models. Reason better aligned philosophy inference, as they model reasons up-dating be...

Recent studies suggest that humans can infer the underlying causal model from observing the distribution of variables. In a multiple-cue experiment we investigated if people can infer the causal structure from mere observation, and if different causal models invite different cognitive processes. Participants performed 220 training trials in two judgment tasks with different underlying causal st...

Kang, Changsung, "Model testing for causal models" (2008). Graduate Theses and Dissertations. Paper 11145.

This chapter reviews the diverse roles that causal knowledge plays in reinforcement learning. The first half of the chapter contrasts a “model-free” system that learns to repeat actions that lead to reward with a “model-based” system that learns a probabilistic causal model of the environment which it then uses to plan action sequences. Evidence suggests that these two systems coexist in the br...

Decision support systems may require the use of existing complex mathematical models. It is desirable to reduce the equations of such a model to an explanatory causal form to support decision making, for example in making decisions about the greenhouse effect. This paper presents an approach to constructing causal explanation from existing complex mathematical models. The method combines inform...

Learning to understand a single causal system can be an achievement, but humans must learn about multiple causal systems over the course of a lifetime. We present a hierarchical Bayesian framework that helps to explain how learning about several causal systems can accelerate learning about systems that are subsequently encountered. Given experience with a set of objects, our framework learns a ...

Computational models of analogical inference have assumed that inferences about the target are generated solely by “copying with substitution and generation” from the source, guided by a mapping based on similarity and parallel structure. In contrast, work in philosophy of science has stressed that analogical inference is based on causal models of the source and target. In two experiments, we s...