Causal Discovery in Manufacturing: A Structured Literature Review

Industry 4.0 radically alters manufacturing organization and management, fostering collection and analysis of increasing amounts of data. Advanced data analytics, such as machine learning (ML), are essential for implementing Industry 4.0 and obtaining insights regarding production, better decision support, and enhanced manufacturing quality and sustainability. ML outperforms traditional approaches in many cases, but its complexity leads to unclear bases for decisions. Thus, acceptance of ML and, concomitantly, Industry 4.0, is hindered due to increasing requirements of fairness, accountability, and transparency, especially in sensitive-use cases. ML does not augment organizational knowledge, which is highly desired by manufacturing experts. Causal discovery promises a solution by providing insights on causal relationships that go beyond traditional ML’s statistical dependency. Causal discovery has a theoretical background and been successfully applied in medicine, genetics, and ecology. However, in manufacturing, only experimental and scattered applications are known; no comprehensive overview about how causal discovery can be applied in manufacturing is available. This paper investigates the state and development of research on causal discovery in manufacturing by focusing on motivations for application, common application scenarios and approaches, impacts, and implementation challenges. Based on the structured literature review, four core areas are identified, and a research agenda is proposed.

The Brain Is Globally Symmetric: An Analysis of Intra and Interhemispheric Symmetry of Subcortical Structures in the Aging Human Brain

Here we address the hemispheric interdependency of subcortical structures in the aging human brain. In particular, we investigate whether volume variation can be explained with the adjacency of structures in the same hemisphere or is due to the interhemispheric development of mirror subcortical structures in the brain. Seven subcortical structures in both hemispheres were automatically segmented in a large sample of over three 3,312 magnetic resonance imaging (MRI) studies of elderly individuals in their 70s and 80s. We perform Eigenvalue analysis to find that anatomic volumes in the limbic system and basal ganglia show similar statistical dependency when considered in the same hemisphere (intrahemispheric) or in different hemispheres (interhemispheric). Our results indicate that anatomic bilaterality is preserved in the aging human brain, supporting the hypothesis that coupling between non-adjacent brain areas could act as a mechanism to compensate for the deleterious effects of aging.

Quantifying the dependence of electron fluxes in the Earth’s radiation belts with radial diffusion drivers through the use of information theory.

<p>We describe the use of information-theoretic methodologies to characterise statistical dependencies of energetic electron fluxes (130 keV and >1 MeV) with a wide range of solar wind and magnetospheric drivers. We focus specifically on drivers associated with radial diffusion processes and revisit the events studied by Rostoker et al. <em>Geophys. Res. Lett.</em> (1998) in terms of mutual information. The main benefit of mutual information, in comparison to the Pearson correlation and other linear measures, lies in the capacity to distinguish nonlinear statistical dependencies from linear ones.  We find that observed enhancement in relativistic electron fluxes correlate weakly, both linearly and nonlinearly, with the ULF power spectrum, whereas less energetic electron fluxes show stronger statistical dependency with both ground and <em>in situ </em>ULF wave power. Our results are indicative of the need to incorporate data analysis tools that can distinguish between interdependencies of various solar wind drivers.</p>

The Formation and Retrieval of Holistic Event Memories Across Development

Event memories consist of associations between their constituent elements, leading to their holistic retrieval via the process of pattern completion. This holistic retrieval can occur, under specific conditions, when each within-event association is encoded in a separate temporal context: adults are able to integrate the information into a single coherent representation. In this study, we sought to replicate the holistic retrieval of simultaneously encoded event elements in children, and examine whether children can similarly integrate across separated encoding contexts. Children (aged 6-7 years; 9-10 years) and adults encoded two series of three-element “events” consisting of an animal, object, and location. In the simultaneous condition, they encountered all three event elements at once; in the separated condition, they encountered each pairwise association separately (animal-object, animal-location, object-location). After encoding, they were tested on the retrieval of each within-event association using a 4-alternative-forced-choice task. We inferred the presence of holistic retrieval using a measure of retrieval dependency—the statistical dependency between retrieval of within-event associations. Memory for the pairs improved across ages, but there were no developmental differences in retrieval dependency. In the simultaneous encoding condition, all three age groups showed retrieval dependency. However, counter to previous studies, retrieval dependency was not observed in any age group following separated encoding. The results from the simultaneous encoding condition support the idea that pattern completion processes are developed by early childhood. The absence of retrieval dependency in adults following separated encoding prevent conclusions regarding the developmental trajectory of mnemonic integration.

Measuring Independence between Statistical Randomness Tests by Mutual Information

The analysis of independence between statistical randomness tests has had great attention in the literature recently. Dependency detection between statistical randomness tests allows one to discriminate statistical randomness tests that measure similar characteristics, and thus minimize the amount of statistical randomness tests that need to be used. In this work, a method for detecting statistical dependency by using mutual information is proposed. The main advantage of using mutual information is its ability to detect nonlinear correlations, which cannot be detected by the linear correlation coefficient used in previous work. This method analyzes the correlation between the battery tests of the National Institute of Standards and Technology, used as a standard in the evaluation of randomness. The results of the experiments show the existence of statistical dependencies between the tests that have not been previously detected.

Sex-specific relationships between face memory and the N170 component in event-related potentials

At the group level, women consistently perform better in face memory tasks than men and also show earlier and larger N170 components of event-related brain potentials (ERP), considered to indicate perceptual structural encoding of faces. Here we investigated sex differences in the relationship between the N170 and face memory performance in 152 men and 141 women at group mean and individual differences levels. ERPs and performance were measured in separate tasks, avoiding statistical dependency between the two. We confirmed previous findings about superior face memory in women and a—sex-independent—negative relationship between N170 latency and face memory. However, whereas in men, better face memory was related to larger N170 components, face memory in women was unrelated with the amplitude or latency of the N170. These data provide solid evidence that individual differences in face memory within men are at least partially related to more intense structural face encoding.

On a Class of Tensor Markov Fields

Here, we introduce a class of Tensor Markov Fields intended as probabilistic graphical models from random variables spanned over multiplexed contexts. These fields are an extension of Markov Random Fields for tensor-valued random variables. By extending the results of Dobruschin, Hammersley and Clifford to such tensor valued fields, we proved that tensor Markov fields are indeed Gibbs fields, whenever strictly positive probability measures are considered. Hence, there is a direct relationship with many results from theoretical statistical mechanics. We showed how this class of Markov fields it can be built based on a statistical dependency structures inferred on information theoretical grounds over empirical data. Thus, aside from purely theoretical interest, the Tensor Markov Fields described here may be useful for mathematical modeling and data analysis due to their intrinsic simplicity and generality.

A Causal Inference Method for Reducing Gender Bias in Word Embedding Relations

Word embedding has become essential for natural language processing as it boosts empirical performances of various tasks. However, recent research discovers that gender bias is incorporated in neural word embeddings, and downstream tasks that rely on these biased word vectors also produce gender-biased results. While some word-embedding gender-debiasing methods have been developed, these methods mainly focus on reducing gender bias associated with gender direction and fail to reduce the gender bias presented in word embedding relations. In this paper, we design a causal and simple approach for mitigating gender bias in word vector relation by utilizing the statistical dependency between gender-definition word embeddings and gender-biased word embeddings. Our method attains state-of-the-art results on gender-debiasing tasks, lexical- and sentence-level evaluation tasks, and downstream coreference resolution tasks.

The Method of Statistical Dependencies of Equations: Application Criteria and Main Functions

The objective assessment of the state and development of socio-economic phenomena and the processes can only be ensured by the use of properly selected statistical and mathematical methods. These methods cease to be a matter of interest in practice when there is no certainty as to how much they can be applied to specific tasks. In the context of significant advances in technology for the collection and processing of statistics, the skilled use of statistical and mathematical methods is significantly hampered by a lack of knowledge of methods and their ability to evaluate information. The article highlights the criteria for choosing the best equation of dependencies as a means of statistical analysis of the influence of factors on the results of socio-economic development of the national economy as a whole, including individual types or forms of economic activity. It has been shown that the statistical dependency method can be used to solve the tasks set out in the article. The method of statistical dependence equations is a statistical method of analyzing the causal relationships of economic phenomena and processes. Unlike the mathematical method of correlation and regression analysis, which is based on linear algebra, the application of the method of statistical equations of dependencies is based on the calculation of the coefficients of comparison, which are determined by the ratio of the individual values of the eponymous sign to its minimum or maximum level. With increasing values of the sign, the coefficients of comparison are calculated from the minimum level, and at decrease — from the maximum. The comparison coefficients show the degree of change (increase or decrease) of the magnitude of the trait to the accepted comparison base. The parameter of the equation of dependence is calculated on the basis of the coefficients of comparison of the resultant and factor trait. Unlike the coefficients of elasticity known in statistics, the parameters of the equation of dependence, which are determined by the method of deviations, allow to take into account not only the influence on the effective attribute of one factor, but also the cumulative effect of many factors.