Causal Feature Selection with Missing Data

Causal feature selection aims at learning the Markov blanket (MB) of a class variable for feature selection. The MB of a class variable implies the local causal structure among the class variable and its MB and all other features are probabilistically independent of the class variable conditioning on its MB, this enables causal feature selection to identify potential causal features for feature selection for building robust and physically meaningful prediction models. Missing data, ubiquitous in many real-world applications, remain an open research problem in causal feature selection due to its technical complexity. In this article, we discuss a novel multiple imputation MB (MimMB) framework for causal feature selection with missing data. MimMB integrates Data Imputation with MB Learning in a unified framework to enable the two key components to engage with each other. MB Learning enables Data Imputation in a potentially causal feature space for achieving accurate data imputation, while accurate Data Imputation helps MB Learning identify a reliable MB of the class variable in turn. Then, we further design an enhanced kNN estimator for imputing missing values and instantiate the MimMB. In our comprehensively experimental evaluation, our new approach can effectively learn the MB of a given variable in a Bayesian network and outperforms other rival algorithms using synthetic and real-world datasets.

Experimental entanglement of temporal order

The study of causal relations has recently been applied to the quantum realm, leading to the discovery that not all physical processes have a definite causal structure. While indefinite causal processes have previously been experimentally shown, these proofs relied on the quantum description of the experiments. Yet, the same experimental data could also be compatible with definite causal structures within different descriptions. Here, we present the first demonstration of indefinite temporal order outside of quantum formalism. We show that our experimental outcomes are incompatible with a class of generalised probabilistic theories satisfying the assumptions of locality and definite temporal order. To this end, we derive physical constraints (in the form of a Bell-like inequality) on experimental outcomes within such a class of theories. We then experimentally invalidate these theories by violating the inequality using entangled temporal order. This provides experimental evidence that there exist correlations in nature which are incompatible with the assumptions of locality and definite temporal order.

Theoretical Notes on Action Schemes in Society

This essay presents theoretical constructs for characterizing the causal structure of social actions and developing a multi-level theory of action relating to accomplishment of goals via social organizations. Focal concepts include: action schemes, mobilization, internal and external fulfillments, power schemes, macroactions, effective actions, and purposeful actions. Additionally, an overview is provided of a methodological procedure for analyzing narratives in order to specify causal linkages among actions and thereby delineate action schemes. Some possibilities for future developments are noted.

Optimal Neighborhood Selection for AR-ARCH Random Fields with Application to Mortality

This article proposes an optimal and robust methodology for model selection. The model of interest is a parsimonious alternative framework for modeling the stochastic dynamics of mortality improvement rates introduced recently in the literature. The approach models mortality improvements using a random field specification with a given causal structure instead of the commonly used factor-based decomposition framework. It captures some well-documented stylized facts of mortality behavior including: dependencies among adjacent cohorts, the cohort effects, cross-generation correlations, and the conditional heteroskedasticity of mortality. Such a class of models is a generalization of the now widely used AR-ARCH models for univariate processes. A the framework is general, it was investigated and illustrated a simple variant called the three-level memory model. However, it is not clear which is the best parameterization to use for specific mortality uses. In this paper, we investigate the optimal model choice and parameter selection among potential and candidate models. More formally, we propose a methodology well-suited to such a random field able to select thebest model in the sense that the model is not only correct but also most economical among all thecorrectmodels. Formally, we show that a criterion based on a penalization of the log-likelihood, e.g., the using of the Bayesian Information Criterion, is consistent. Finally, we investigate the methodology based on Monte-Carlo experiments as well as real-world datasets.

Identifying Causes of Social Evolution: Contextual Analysis, the Price Approach, and Multilevel Selection

Kin selection theory and multilevel selection theory are distinct approaches to explaining the evolution of social traits. The latter claims that it is useful to regard selection as a process that can occur on multiple levels of organisation such as the level of individuals and the level of groups. This is reflected in a decomposition of fitness into an individual component and a group component. This multilevel view is central to understanding and characterising evolutionary transitions in individuality, e.g., from unicellular life to multicellular organisms, but currently suffers from the lack of a consistent, quantifiable measure. Specifically, the two major statistical tools to determine the coefficients of such a decomposition, the multilevel Price equation and contextual analysis, are inconsistent and may disagree on whether group selection is present. Here we show that the reason for the discrepancies is that underlying the multilevel Price equation and contextual analysis are two non-equivalent causal models for the generation of individual fitness effects (thus leaving different “remainders” explained by group effects). While the multilevel Price equation assumes that the individual effect of a trait determines an individual's relative success within a group, contextual analysis posits that the individual effect is context-independent. Since these different assumptions reflect claims about the causal structure of the system, the correct approach cannot be determined on general theoretical or statistical grounds but must be identified by experimental intervention. We outline interventions that reveal the underlying causal structure and thus facilitate choosing the appropriate approach. We note that kin selection theory with its focus on the individual is immune to such inconsistency because it does not address causal structure with respect to levels of organisation. In contrast, our analysis of the two approaches to measuring group selection demonstrates that multilevel selection theory adds meaningful (falsifiable) causal structure to explain the sources of individual fitness and thereby constitutes a proper refinement of kin selection theory. Taking such refined causal structure into account seems indispensable for studying evolutionary transitions in individuality because these transitions are characterised by changes in the selection pressures that act on the respective levels.

Neural dynamics of causal inference in the macaque frontoparietal circuit

ABSTRACTNatural perception relies inherently on inferring causal structure in the environment. However, the neural mechanisms and functional circuits that are essential for representing and updating the hidden causal structure and corresponding sensory representations during multisensory processing are unknown. To address this, monkeys were trained to infer the probability of a potential common source from visual and proprioceptive signals on the basis of their spatial disparity in a virtual reality system. The proprioceptive drift reported by monkeys demonstrated that they combined historical information and current multisensory signals to estimate the hidden common source and subsequently updated both the causal structure and sensory representation. Single-unit recordings in premotor and parietal cortices revealed that neural activity in premotor cortex represents the core computation of causal inference, characterizing the estimation and update of the likelihood of integrating multiple sensory inputs at a trial-by-trial level. In response to signals from premotor cortex, neural activity in parietal cortex also represents the causal structure and further dynamically updates the sensory representation to maintain consistency with the causal inference structure. Thus, our results indicate how premotor cortex integrates historical information and sensory inputs to infer hidden variables and selectively updates sensory representations in parietal cortex to support behavior. This dynamic loop of frontal-parietal interactions in the causal inference framework may provide the neural mechanism to answer long-standing questions regarding how neural circuits represent hidden structures for body-awareness and agency.

Understanding the Performance of Construction Business: A Simulation-Based Experimental Study

Higher failure rates of construction business have been observed as a recurring phenomenon in the construction industry. This research focuses on the causes behind a range of performance modes of construction business. The growth and capacity under-investment archetype has been used as the main systems archetype to develop a causal structure for understanding the business performance. A system dynamics model was developed to create a simulation platform for the causal structure. A context of a typical small and medium construction company has been used in the simulation model. This research considered and experimented with a set of selected managerial policies and practices that can lead the construction business to failure, sustenance, or growth. In order to achieve the expected growth or sustenance, it is found that a certain level of balance needs to be secured on how much emphasis is to be given to win new projects, how much profit margins to work with, and how much capacities to be arranged and deployed for project operations, management, and execution.