Directed acyclic graphical (DAG) models are a powerful tool for representing causal relationships among jointly distributed random variables, especially concerning data from across different experimental settings. However, it is not always practical or desirable to estimate a causal model at the granularity of given features in a particular dataset. There is a growing body of research on causal abstraction to address such problems. We contribute to this line of research by (i) providing novel graphical identifiability results for practically-relevant interventional settings, (ii) proposing an efficient, provably consistent algorithm for directly learning abstract causal graphs from interventional data with unknown intervention targets, and (iii) uncovering theoretical insights about the lattice structure of the underlying search space, with connections to the field of causal discovery more generally. As proof of concept, we apply our algorithm on synthetic and real datasets with known ground truths, including measurements from a controlled physical system with interacting light intensity and polarization.

Recent work on causal abstraction, in particular graphical approaches focusing on causal structure between clusters of variables, aims to summarize a high-dimensional causal structure in terms of a low-dimensional one. Existing methods for learning such summaries from data assume that both the high- and low-dimensional structures are acyclic, which is helpful for causal effect identification and reasoning but excludes many high-dimensional models and thus limits applicability. We show that in the linear non-Gaussian (LiNG) setting, the high-dimensional acyclicity assumption can be relaxed while still allowing recovery of a low-dimensional causal directed acyclic graph (DAG). We further connect identifiability of this low-dimensional DAG to existing results: LiNG models with cycles are observationally identifiable only up to an equivalence class whose members differ by reversals of directed cycles; our low-dimensional DAG, which is invariant across all members of a given equivalence class, thus forms a natural representative of the class. While existing approaches for learning this observational equivalence class over high-dimensional variables have exponential time complexity, our low-dimensional summary is learned in worst-case cubic time and comes with explicit bounds on the sample complexity. We provide open source code and experiments on synthetic data to corroborate our theoretical results.

In designing generative models, it is commonly believed that in order to learn useful latent structure, we face a fundamental tension between expressivity and structure. In this paper we challenge this view by proposing a new approach to training arbitrarily expressive generative models that simultaneously learn causally disentangled concepts. This is accomplished by adding a simple context module to an arbitrarily complex black-box model, which learns to process concept information by implicitly inverting linear representations from the model's encoder. Inspired by the notion of intervention in a causal model, our module selectively modifies its architecture during training, allowing it to learn a compact joint model over different contexts. We show how adding this module leads to causally disentangled representations that can be composed for out-of-distribution generation on both real and simulated data. The resulting models can be trained end-to-end or fine-tuned from pre-trained models. To further validate our proposed approach, we prove a new identifiability result that extends existing work on identifying structured representations.

Behaviour modelling has been widely explored using both statistical and machine learning techniques, primarily relying on analyzing correlations to understand passenger responses under different conditions and scenarios. However, correlation alone does not imply causation. This paper introduces a data-driven causal behaviour modelling approach, comprising two phases: causal discovery and causal inference. Causal discovery phase uses Peter-Clark (PC) algorithm to learn a directed acyclic graph that captures the causal relationships among variables. Causal inference phase estimates the corresponding model parameters and infers (conditional) causal effects of interventions designed to influence user behaviour. The method is validated by comparing the results with those from conventional modelling approaches (logistic regression and expert knowledge) using smart card data from a real-world use case on a pre-peak fare discount incentive program in the Hong Kong Mass Transit Railway system. The results highlight that the purely data-driven causal discovery method can produce reasonable causal graph. The method can also quantify the behavioural impacts of the incentive, identify key influencing factors, and estimate the corresponding causal effects. The overall causal effect of the incentive is approximately 0.7 %, with about 3 % of the population changing behaviour from previous statistical analysis. Interestingly, passengers with the highest flexibility exhibit a negative response, while those with medium-to-high flexibility demonstrate 3 times of the general level of responsiveness. The approach initiates the data-driven, causal modelling of human behaviour dynamics to support policy developments and managerial interventions.

Factor analysis (FA) is a statistical tool for studying how observed variables with some mutual dependences can be expressed as functions of mutually independent unobserved factors, and it is widely applied throughout the psychological, biological, and physical sciences. We revisit this classic method from the comparatively new perspective given by advancements in causal discovery and deep learning, introducing a framework for Neuro-Causal Factor Analysis (NCFA). Our approach is fully nonparametric: it identifies factors via latent causal discovery methods and then uses a variational autoencoder (VAE) that is constrained to abide by the Markov factorization of the distribution with respect to the learned graph. We evaluate NCFA on real and synthetic data sets, finding that it performs comparably to standard VAEs on data reconstruction tasks but with the advantages of sparser architecture, lower model complexity, and causal interpretability. Unlike traditional FA methods, our proposed NCFA method allows learning and reasoning about the latent factors underlying observed data from a justifiably causal perspective, even when the relations between factors and measurements are highly nonlinear.

Several approaches to graphically representing context-specific relations among jointly distributed categorical variables have been proposed, along with structure learning algorithms. While existing optimization-based methods have limited scalability due to the large number of context-specific models, the constraint-based methods are more prone to error than even constraint-based directed acyclic graph learning algorithms since more relations must be tested. We present an algorithm for learning context-specific models that scales to hundreds of variables. Scalable learning is achieved through a combination of an order-based Markov chain Monte-Carlo search and a novel, context-specific sparsity assumption that is analogous to those typically invoked for directed acyclic graphical models. Unlike previous Markov chain Monte-Carlo search methods, our Markov chain is guaranteed to have the true posterior of the variable orderings as the stationary distribution. To implement the method, we solve a first case of an open problem recently posed by Alon and Balogh. Future work solving increasingly general instances of this problem would allow our methods to learn increasingly dense models. The method is shown to perform well on synthetic data and real world examples, in terms of both accuracy and scalability.

We consider the problem of estimating the marginal independence structure of a Bayesian network from observational data, learning an undirected graph we call the unconditional dependence graph. We show that unconditional dependence graphs of Bayesian networks correspond to the graphs having equal independence and intersection numbers. Using this observation, a Gr\"obner basis for a toric ideal associated to unconditional dependence graphs of Bayesian networks is given and then extended by additional binomial relations to connect the space of all such graphs. An MCMC method, called GrUES (Gr\"obner-based Unconditional Equivalence Search), is implemented based on the resulting moves and applied to synthetic Gaussian data. GrUES recovers the true marginal independence structure via a penalized maximum likelihood or MAP estimate at a higher rate than simple independence tests while also yielding an estimate of the posterior, for which the $20\%$ HPD credible sets include the true structure at a high rate for data-generating graphs with density at least $0.5$.

We present Queer in AI as a case study for community-led participatory design in AI. We examine how participatory design and intersectional tenets started and shaped this community's programs over the years. We discuss different challenges that emerged in the process, look at ways this organization has fallen short of operationalizing participatory and intersectional principles, and then assess the organization's impact. Queer in AI provides important lessons and insights for practitioners and theorists of participatory methods broadly through its rejection of hierarchy in favor of decentralization, success at building aid and programs by and for the queer community, and effort to change actors and institutions outside of the queer community. Finally, we theorize how communities like Queer in AI contribute to the participatory design in AI more broadly by fostering cultures of participation in AI, welcoming and empowering marginalized participants, critiquing poor or exploitative participatory practices, and bringing participation to institutions outside of individual research projects. Queer in AI's work serves as a case study of grassroots activism and participatory methods within AI, demonstrating the potential of community-led participatory methods and intersectional praxis, while also providing challenges, case studies, and nuanced insights to researchers developing and using participatory methods.

We consider the problem of characterizing directed acyclic graph (DAG) models up to unconditional equivalence, i.e., when two DAGs have the same set of unconditional d-separation statements. Each unconditional equivalence class (UEC) can be uniquely represented with an undirected graph whose clique structure encodes the members of the class. Via this structure, we provide a transformational characterization of unconditional equivalence. Combining these results, we introduce a hybrid algorithm for learning DAG models from observational data, called Greedy Unconditional Equivalence Search (GUES), which first estimates the UEC of the data using independence tests and then greedily searches the UEC for the optimal DAG. Applying GUES on synthetic data, we show that it achieves comparable accuracy to existing methods. However, in contrast to existing methods, since the average UEC is observed to contain few DAGs, the search space for GUES is drastically reduced.

We consider the problem of characterizing Bayesian networks up to unconditional equivalence, i.e., when directed acyclic graphs (DAGs) have the same set of unconditional $d$-separation statements. Each unconditional equivalence class (UEC) is uniquely represented with an undirected graph whose clique structure encodes the members of the class. Via this structure, we provide a transformational characterization of unconditional equivalence; i.e., we show that two DAGs are in the same UEC if and only if one can be transformed into the other via a finite sequence of specified moves. We also extend this characterization to the essential graphs representing the Markov equivalence classes (MECs) in the UEC. UECs partition the space of MECs and are easily estimable from marginal independence tests. Thus, a characterization of unconditional equivalence has applications in methods that involve searching the space of MECs of Bayesian networks.

Riemannian frameworks are the basis for some of the best-performing decoding methods in EEG-based Brain-Computer Interfacing. In this work, we consider whether a nonlinear extension of the Riemannian framework, obtained by replacing the channel-wise covariance of the EEG signal with the nonlinear distance covariance, improves decoding performance. We study the theoretical properties of the distance covariance metric in this framework, in particular invariance to affine transformations, and compare the proposed method with established Riemannian methods on three different EEG data sets. We do not find evidence that the distance covariance extension improves decoding performance in comparison to the linear Riemannian framework.

Author(s): Meunier, Anja; Markham, Alex; Grosse-Wentrup, Moritz | Abstract: The question of how algorithms in general and cognitive skills in particular are implemented by our nervous system is at the core of cognitive science. The notions of what it means for a physical system (such as our nervous system) to implement an algorithm, however, are surprisingly vague. We argue that a rigorous theory is needed to formulate and evaluate precise hypotheses about the brain's cognitive functions and propose a definition of the term algorithm as a chain of functions. Subsequently, we define the term implementation via a sequence of projections from a dynamical system, represented by a Markov process, to the algorithm. We furthermore show the practical applicability of this approach in a simulated example. We believe that the theory proposed here contributes to bridging the gap between the algorithmic and the implementational level by rendering the task at hand theoretically precise.

We consider the problem of causal structure learning in the setting of heterogeneous populations, i.e., populations in which a single causal structure does not adequately represent all population members, as is common in biological and social sciences. To this end, we introduce a distance covariance-based kernel designed specifically to measure the similarity between the underlying nonlinear causal structures of different samples. Indeed, we prove that the corresponding feature map is a statistically consistent estimator of nonlinear independence structure, rendering the kernel itself a statistical test for the hypothesis that sets of samples come from different generating causal structures. Even stronger, we prove that the kernel space is isometric to the space of causal ancestral graphs, so that distance between samples in the kernel space is guaranteed to correspond to distance between their generating causal structures. This kernel thus enables us to perform clustering to identify the homogeneous subpopulations, for which we can then learn causal structures using existing methods. Though we focus on the theoretical aspects of the kernel, we also evaluate its performance on synthetic data and demonstrate its use on a real gene expression data set.

We consider the task of causal structure learning over measurement dependence inducing latent (MeDIL) causal models. We show that this task can be framed in terms of the graph theoretic problem of finding edge clique covers,resulting in an algorithm for returning minimal MeDIL causal models (minMCMs). This algorithm is non-parametric, requiring no assumptions about linearity or Gaussianity. Furthermore, despite rather weak assumptions aboutthe class of MeDIL causal models, we show that minimality in minMCMs implies some rather specific and interesting properties. By establishing MeDIL causal models as a semantics for edge clique covers, we also provide a starting point for future work further connecting causal structure learning to developments in graph theory and network science.

We present the MeDIL Python package for causal modelling. Its current features focus on (i) nonlinear unconditional pairwise independence testing, (ii) constraint-based causal structure learning, and (iii) learning the corresponding functional causal models (FCMs), all for the class of measurement dependence inducing latent (MeDIL) causal models. MeDIL causal models and therefore the MeDIL software package are especially suited for analyzing data from fields such as psychometric, epidemiology, etc. that rely on questionnaire or survey data.