Causal Independence Research Articles

CIS2N: Causal independence and sparse shift network for rotating machinery fault diagnosis in unseen domains

Intelligent fault diagnosis (IFD) based on deep learning (DL) has demonstrated its powerful performance to promote the reliability and safe operation of rotating machinery. In industrial applications, working condition changes frequently, leading to models’ performance degeneration due to out-of-distribution (OOD) samples, such as vibration signals with unseen rotating speeds, loads, etc. To address this issue, domain generalization fault diagnosis (DGFD) has attracted increasing attention. However, mainstream DGFD methods are implemented from statical dependence, which fail to excavate intrinsic causal mechanisms and have limited performance and reliability. From a causal view, the causal independence and sparse shift network (CIS2N) is proposed to diagnose via causal mechanisms. Specially, a structural causal model (SCM) representing causal relations of features in IFD is given. Based on the SCM, it is derived that features are jointly independent and feature deviation between intervened sample pairs is sparse. The dependence among features is measured via the covariance matrices constructed by high-level features mapped by random Fourier features (RFF). Meanwhile the L1-distance between pairs of features with the same causal features (intervened pair) is calculated and optimized to realize shift sparsity. The effectiveness of the proposed CIS2N on DGFD is demonstrated on two real helical gearbox datasets.

Causality Investigation between Gut Microbiota, Derived Metabolites, and Obstructive Sleep Apnea: A Bidirectional Mendelian Randomization Study.

Various studies have highlighted the important associations between obstructive sleep apnea (OSA) and gut microbiota and related metabolites. Nevertheless, the establishment of causal relationships between these associations remains to be determined. Multiple mendelian randomization (MR) analyses were performed to genetically predict the causative impact of 196 gut microbiota and 83 metabolites on OSA. Two-sample MR was used to assess the potential association, and causality was evaluated using inverse variance weighted (IVW), MR-Egger, and weighted median (WM) methods. Multivariable MR (MVMR) was employed to ascertain the causal independence between gut microbiota and the metabolites linked to OSA. Additionally, Cochran's Q test, the MR Egger intercept test and the MR Steiger test were used for the sensitivity analyses. The analysis of the 196 gut microbiota revealed that genus_Ruminococcaceae (UCG009) (PIVW = 0.010) and genus_Subdoligranulum (PIVW = 0.041) were associated with an increased risk of OSA onset. Conversely, Family_Ruminococcaceae (PIVW = 0.030), genus_Coprococcus2 (PWM = 0.025), genus_Eggerthella (PIVW = 0.011), and genus_Eubacterium (xylanophilum_group) (PIVW = 0.001) were negatively related to the risk of OSA. Among the 83 metabolites evaluated, 3-dehydrocarnitine, epiandrosterone sulfate, and leucine were determined to be potential independent risk factors associated with OSA. Moreover, the reverse MR analysis demonstrated a suggestive association between OSA exposure and six microbiota taxa. This study offers compelling evidence regarding the potential beneficial or detrimental causative impact of the gut microbiota and its associated metabolites on OSA risk, thereby providing new insights into the mechanisms of gut microbiome-mediated OSA development.

Leibniz’s Causal Road to Existential Independence

Abstract Leibniz thinks that every created substance is causally active, and yet causally independent of every other: none can cause changes in any but itself. This is not controversial. But Leibniz also thinks that every created substance is existentially independent of every other: it is metaphysically possible for any to exist with or without any other. This is controversial. I argue that, given a mainstream reading of Leibniz’s essentialism, if one accepts the former, uncontroversial interpretation concerning causal independence, then one ought also to accept the latter, controversial one concerning existential independence. This is a new way to defend the ‘existential independence’ interpretation. Moreover, this defense provides a new approach for defending the broadly ‘non-logical’ interpretive camp in the longstanding debate over Leibniz’s views on incompossibility, against perhaps the strongest objection leveled by advocates of the opposing broadly ‘logical’ interpretation.

Against the Flow of Time with Multi-Output Models

Abstract Recent work has paid close attention to the first principle of Granger causality, according to which cause precedes effect. In this context, the question may arise whether the detected direction of causality also reverses after the time reversal of unidirectionally coupled data. Recently, it has been shown that for unidirectionally causally connected autoregressive (AR) processes X → Y, after time reversal of data, the opposite causal direction Y → X is indeed detected, although typically as part of the bidirectional X ↔ Y link. As we argue here, the answer is different when the measured data are not from AR processes but from linked deterministic systems. When the goal is the usual forward data analysis, cross-mapping-like approaches correctly detect X → Y, while Granger causality-like approaches, which should not be used for deterministic time series, detect causal independence X ⫫ Y . The results of backward causal analysis depend on the predictability of the reversed data. Unlike AR processes, observables from deterministic dynamical systems, even complex nonlinear ones, can be predicted well forward, while backward predictions can be difficult (notably when the time reversal of a function leads to one-to-many relations). To address this problem, we propose an approach based on models that provide multiple candidate predictions for the target, combined with a loss function that consideres only the best candidate. The resulting good forward and backward predictability supports the view that unidirectionally causally linked deterministic dynamical systems X → Y can be expected to detect the same link both before and after time reversal.

Tossing Morgenbesser’s Coin

Abstract Morgenbesser's Coin is a thought experiment that exemplifies a widespread disposition to infer counterfactual independence from causal independence. I argue that this disposition is mistaken by analysing a closely related thought experiment.

Quantum teleportation and entanglement swapping with long baseline in outer space

Quantum information (QI) experiments applying quantum optics in outer space with a very long baseline may have advantages over the current Earth-bound experiments or the Earth-to-satellite experiments, because they can minimize the loss in light transmission and maximize the gain in time resolution. This future class of experiments, among them quantum teleportation and entanglement swapping, can shed light on many fundamental theoretical issues in gravitational quantum physics and relativistic QI. Regarding relativity theory, these experiments in an outer-space setting can involve observations at spacelike and timelike separations and explicate intriguing phenomena from different choices of time-slicing. Regarding QI, they may be able to ensure the causal independence of the expectation values in the Bell test. These issues are addressed in this paper with analysis and explanations.

Morgenbesser’s Coin

AbstractBefore a fair, indeterministic coin is tossed, Lucky, who is causally isolated from the coin-tossing mechanism, declines to bet on heads. The coin lands heads. The consensus is that the following counterfactual is true:(M:) If Lucky had bet heads, he would have won the bet.It is also widely believed that to rule (M) true, any plausible semantics for counterfactuals must invoke causal independence. But if that’s so, the hope of giving a reductive analysis of causation in terms of counterfactuals is undermined. Here I argue that there is compelling reason to question the assumption that (M) is true.

The freedom we mean: A causal independence account of creativity and academic freedom

Academic freedom has often been defended in a progressivist manner: without academic freedom, creativity would be in peril, and with it the advancement of knowledge, i.e. the epistemic progress in science. In this paper, I want to critically discuss the limits of such a progressivist defense of academic freedom, also known under the label ‘argument from truth.’ The critique is offered, however, with a constructive goal in mind, namely to offer an alternative account that connects creativity and academic freedom in a way that goes beyond mere reference to epistemic progress and involves reference to the freedom to think independently as the freedom we mean when we point to creativity and when we point to academic freedom. The resulting causal independence account is not only epistemologically stronger than a progressivist account, it also allows to counter the curbing of academic freedom in the name of progress. The latter becomes key, for instance, when authoritarian political regimes limit or negate academic freedom with reference to an epistemic progress suitably defined for that regime.

Efficient Decomposition of Bayesian Networks With Non-graded Variables

Elicitation, estimation and exact inference in Bayesian Networks (BNs) are often difficult because the dimension of each Conditional Probability Table (CPT) grows exponentially with the increase in the number of parent variables. The Noisy-MAX decomposition has been proposed to break down a large CPT into several smaller CPTs exploiting the assumption of causal independence, i.e., absence of causal interaction among parent variables. In this way, the number of conditional probabilities to be elicited or estimated and the computational burden of the joint tree algorithm for exact inference are reduced. Unfortunately, the Noisy-MAX decomposition is suited to graded variables only, i.e., ordinal variables with the lowest state as reference, but real-world applications of BNs may also involve a number of non-graded variables, like the ones with reference state in the middle of the sample space (double-graded variables) and with two or more unordered non-reference states (multi-valued nominal variables). In this paper, we propose the causal independence decomposition, which includes the Noisy-MAX and two generalizations suited to double-graded and multi-valued nominal variables. While the general definition of BN implicitly assumes the presence of all the possible causal interactions, our proposal is based on causal independence, and causal interaction is a feature that can be added upon need. The impact of our proposal is investigated on a published BN for the diagnosis of acute cardiopulmonary diseases.

Authority (I): A Promise Theoretic Formalization

Authority is a central concept in social systems, but it has a variety of meanings. Promise Theory offers a simple formalized understanding of authority, and its origins, as polarization within a network of collaborative interactions. This idealized approximation stands in contrast to the usual deontic view of authority in socio-philosophical literature, and unifies the various interpretations with a single idea. It's shown that the elementary meanings of authority can all be understood as a promise, analogous to that of a `compass direction' within some decision space, with which agents may choose to align voluntarily. Authority is therefore separated from the embodiment by any particular agency or kind of agent, and is closely related to the concept of leadership in management science. Agents may try to impose authoritative directives onto subordinates, but imposition will generally be ineffective, due to their autonomy or causal independence. Stable configurations may be formed from resonant interactions that employ both semantics and dynamics to bind agents. This simple-minded formalization serves as an foundation for later study about the dynamics of authority and derived `power'.

Explaining Away, Augmentation, and the Assumption of Independence.

In reasoning about situations in which several causes lead to a common effect, a much studied and yet still not well-understood inference is that of explaining away. Assuming that the causes contribute independently to the effect, if we learn that the effect is present, then this increases the probability that one or more of the causes are present. But if we then learn that a particular cause is present, this cause “explains” the presence of the effect, and the probabilities of the other causes decrease again. People tend to show this explaining away effect in their probability judgments, but to a lesser extent than predicted by the causal structure of the situation. We investigated further the conditions under which explaining away is observed. Participants estimated the probability of a cause, given the presence or the absence of another cause, for situations in which the effect was either present or absent, and the evidence about the effect was either certain or uncertain. Responses were compared to predictions obtained using Bayesian network modeling as well as a sensitivity analysis of the size of normative changes in probability under different information conditions. One of the conditions investigated: when there is certainty that the effect is absent, is special because under the assumption of causal independence, the probabilities of the causes remain invariant, that is, there is no normative explaining away or augmentation. This condition is therefore especially diagnostic of people’s reasoning about common-effect structures. The findings suggest that, alongside earlier explanations brought forward in the literature, explaining away may occur less often when the causes are assumed to interact in their contribution to the effect, and when the normative size of the probability change is not large enough to be subjectively meaningful. Further, people struggled when given evidence against negative evidence, resembling a double negation effect.

Trans-Causalizing NAT-Modeled Bayesian Networks.

Conditional independence encoded in Bayesian networks (BNs) avoids combinatorial explosion on the number of variables. However, BNs are still subject to exponential growth of space and inference time on the number of causes per effect variable in conditional probability tables. A number of space-efficient local models exist that allow efficient encoding of dependency between an effect and its causes, and can also be exploited for improved inference efficiency. We focus on the Nonimpeding Noisy-AND Tree (NIN-AND Tree or NAT) models because of multiple merits. We present a novel framework, trans-causalization of NAT-modeled BNs, by which causal independence embedded in NAT models is exploited for more efficient inference. We show that trans-causalization is exact and yields polynomial space complexity. We demonstrate significant efficiency gain on inference based on lazy propagation and sum-product networks.

Counterfactuals and double prevention

Some have argued that no analysis of counterfactual conditionals can succeed without appealing to causal notions. Such authors claim that, in determining what would transpire had some events gone differently, we hold fixed everything that is causally independent from those events. Call this view Causal Independence. Some have argued that we need Causal Independence to accommodate intuitive judgments about certain kinds of counterfactuals in indeterministic worlds. The aim of this paper is to show that, contra these authors, Causal Independence systematically delivers counterintuitive results for a certain subset of such counterfactuals-namely, those involving causation by double prevention. I conclude that intuitions about such counterfactuals do not motivate Causal Independence, at least in any form in which it has thus far been articulated. However, I suggest that a refined Causal Independence thesis that presupposes a kind of causal pluralism might be able to accommodate these intuitions, though such a refined version of Causal Independence may not conflict with reductive analyses of causal notions that appeal to counterfactuals after all.

Causal Inference in Judgment Using the Balanced Scorecard

ABSTRACTOne of the potential threats to the effectiveness of the Balanced Scorecard (BSC) is that managers over- or underuse particular perspectives of the BSC. Specifically, we investigate the effects of (1) the presentation of strategic objectives (generic strategy map versus strategic objective list), and (2) the performance outcome patterns (positive versus negative outer perspective) across the performance measurement perspectives of the BSC and find support that is consistent with the violation of the causal independence assumption (VCIA) in the psychology literature (Rehder 2014). Our findings show that the presentation of the strategic objectives and the performance outcome patterns interact significantly affecting performance evaluation outcomes. Two follow-up experiments provide further support for the VCIA observed in the main experiment by ruling out an alternative explanation that managers simply place a greater emphasis on financial performance measures.Data Availability: Data are available from the authors upon request.

Device-independent quantum secret sharing in arbitrary even dimensions

We present a device independent quantum secret sharing scheme in arbitrary even dimension. We propose a $d$-dimensional $N$-partite linear game, utilizing a generic multipartite higher dimensional Bell inequality, a generalization of Mermin's inequality in the higher dimension. Probability to win this linear game defines the device independence test of the proposed scheme. The security is proved under causal independence of measurement devices and it is based on the polygamy property of entanglement. By defining $\epsilon_{cor}$-correctness and $\epsilon^c$-completeness for a quantum secret sharing scheme, we have also shown that the proposed scheme is $\epsilon_{cor}$-correct and $\epsilon^c$-complete.

Game semantics for quantum programming

Quantum programming languages permit a hardware independent, high-level description of quantum algo rithms. In particular, the quantum lambda-calculus is a higher-order programming language with quantum primitives, mixing quantum data and classical control. Giving satisfactory denotational semantics to the quantum lambda-calculus is a challenging problem that has attracted significant interest in the past few years. Several models have been proposed but for those that address the whole quantum λ-calculus, they either do not represent the dynamics of computation, or they lack the compositionality one often expects from denotational models. In this paper, we give the first compositional and interactive model of the full quantum lambda-calculus, based on game semantics. To achieve this we introduce a model of quantum games and strategies, combining quantum data with a representation of the dynamics of computation inspired from causal models of concurrent systems. In this model we first give a computationally adequate interpretation of the affine fragment. Then, we extend the model with a notion of symmetry, allowing us to deal with replication. In this refined setting, we interpret and prove adequacy for the full quantum lambda-calculus. We do this both from a sequential and a parallel interpretation, the latter representing faithfully the causal independence between sub-computations.

Exploring the Effect of Stimulus Similarity on the Summation Effect in Causal Learning.

Several contemporary models anticipate that the summation effect is modulated by the similarity between the cues forming a compound. Here, we explore this hypothesis in a series of causal learning experiments. Participants were presented with two visual cues that separately predicted a common outcome and later asked for the outcome predicted by the compound of the two cues. Similarity was varied between groups through changes in shape, spatial position, color, configuration, and rotation. In variance with the predictions of these models, we observed similar and strong levels of summation in both groups across all manipulations of similarity. The effect, however, was significantly reduced by manipulations intended to impact assumptions about the causal independence of the cues forming the compound, but this reduction was independent of stimulus similarity. These results are problematic for similarity-based models and can be more readily explained by rational approaches to causal learning.

How Organisms Gained Causal Independence and How It Might Be Quantified.

Two broad features are jointly necessary for autonomous agency: organisational closure and the embodiment of an objective-function providing a ‘goal’: so far only organisms demonstrate both. Organisational closure has been studied (mostly in abstract), especially as cell autopoiesis and the cybernetic principles of autonomy, but the role of an internalised ‘goal’ and how it is instantiated by cell signalling and the functioning of nervous systems has received less attention. Here I add some biological ‘flesh’ to the cybernetic theory and trace the evolutionary development of step-changes in autonomy: (1) homeostasis of organisationally closed systems; (2) perception-action systems; (3) action selection systems; (4) cognitive systems; (5) memory supporting a self-model able to anticipate and evaluate actions and consequences. Each stage is characterised by the number of nested goal-directed control-loops embodied by the organism, summarised as will-nestedness . Organism tegument, receptor/transducer system, mechanisms of cellular and whole-organism re-programming and organisational integration, all contribute to causal independence. Conclusion: organisms are cybernetic phenomena whose identity is created by the information structure of the highest level of causal closure (maximum ), which has increased through evolution, leading to increased causal independence, which might be quantifiable by ‘Integrated Information Theory’ measures.

A multivariate additive noise model for complete causal discovery

Explaining causal reasoning in the form of directed acyclic graphs (DAGs) yields nodal structures with multivariate relationships. In real-world phenomena, these effects can be seen as multiple feature dependency with unmeasured external influences or noises. The bivariate models for causal discovery simply miss to find the multiple feature dependency criteria in the causal models. Here, we propose a multivariate additive noise model (MANM) to solve these issues while analyzing and presenting a multi-nodal causal structure. We introduce new criteria of causal independence for qualitative analysis of causal models and causal influence factor (CIF) for the successful discovery of causal directions in the multivariate system. The scores of CIF provide the information for the goodness of casual inference. The identifiability of the proposed model to discover linear, non-linear causal relations is verified in simulated, real-world datasets and the ability to construct the complete causal model. In comparison test, MANM has out performed Independent Component Analysis based Linear Non-Gaussian Acyclic Model (ICA-LiNGAM), Greedy DAG Search (GDS) and Regression with Sub-sequent Independent Test (RESIT), and performed better for Gaussian and non-Gaussian mixture models with both correlated and uncorrelated feature relations. In performance test, different model fitting errors which occur during causal model construction are discussed and the performance of MANM in comparison to ICA-LiNGAM, GDS and RESIT is provided. Results show that MANM has better causal model construction ability, producing few extra sets of direction with no missing or wrong directions and can estimate every possible causal direction in complex feature sets.

Independence may feel strange

SummaryStochastic independence is not an easy notion. Because it is part of the probability structure, it can have some surprising non‐properties, which is beneficial for teachers and students to see illustrated. Neither is the relationship to causal independence an easy one.