Christian Knoll

Christian Toth, Lars Lorch, Christian Knoll et al. · eth-zurich

Causal discovery and causal reasoning are classically treated as separate and consecutive tasks: one first infers the causal graph, and then uses it to estimate causal effects of interventions. However, such a two-stage approach is uneconomical, especially in terms of actively collected interventional data, since the causal query of interest may not require a fully-specified causal model. From a Bayesian perspective, it is also unnatural, since a causal query (e.g., the causal graph or some causal effect) can be viewed as a latent quantity subject to posterior inference -- other unobserved quantities that are not of direct interest (e.g., the full causal model) ought to be marginalized out in this process and contribute to our epistemic uncertainty. In this work, we propose Active Bayesian Causal Inference (ABCI), a fully-Bayesian active learning framework for integrated causal discovery and reasoning, which jointly infers a posterior over causal models and queries of interest. In our approach to ABCI, we focus on the class of causally-sufficient, nonlinear additive noise models, which we model using Gaussian processes. We sequentially design experiments that are maximally informative about our target causal query, collect the corresponding interventional data, and update our beliefs to choose the next experiment. Through simulations, we demonstrate that our approach is more data-efficient than several baselines that only focus on learning the full causal graph. This allows us to accurately learn downstream causal queries from fewer samples while providing well-calibrated uncertainty estimates for the quantities of interest.

Regina Schuster, Kathleen Gregory, Christian Knoll et al.

Data visualizations are widely used to communicate messages about urgent topics such as climate change and public health. However, we still know little about how these visualizations are produced and interpreted in popular science contexts. In this mixed-method study, we examine how data are visually communicated and understood in the popular science magazine Scientific American, focusing on the messages these visualizations convey. To capture this complexity, we analyze data visualizations about climate change and pandemics in Scientific American over the past fifty years from three complementary perspectives: reader, chart, and producer. From the reader's perspective, we articulate takeaway messages and document sensemaking, interpreting visualizations first without and then with textual elements. From the chart perspective, we examine how visual features and text shape interpretation. From the producer's perspective, we draw on interviews with Scientific American staff to understand message planning and compare a sample of their intended messages with those we interpreted. Using takeaway messages as our central analytic lens, we develop a message typology and show that messages vary systematically across dimensions such as granularity, articulation, and inference. A key finding is that text plays a pivotal role: approximately two-thirds of messages change when textual elements are added. While the interviews highlighted the central role of message planning in visualization production, intended and interpreted messages only partially aligned. Our findings underscore the importance of contextual clarity and audience-aware communication, and we derive recommendations for visualization designers and science communicators.

Christian Knoll

Probabilistic graphical models are a powerful concept for modeling high-dimensional distributions. Besides modeling distributions, probabilistic graphical models also provide an elegant framework for performing statistical inference; because of the high-dimensional nature, however, one must often use approximate methods for this purpose. Belief propagation performs approximate inference, is efficient, and looks back on a long success-story. Yet, in most cases, belief propagation lacks any performance and convergence guarantees. Many realistic problems are presented by graphical models with loops, however, in which case belief propagation is neither guaranteed to provide accurate estimates nor that it converges at all. This thesis investigates how the model parameters influence the performance of belief propagation. We are particularly interested in their influence on (i) the number of fixed points, (ii) the convergence properties, and (iii) the approximation quality.