The Experts below are selected from a list of 33123 Experts worldwide ranked by ideXlab platform
Carol Smidts - One of the best experts on this subject based on the ideXlab platform.
-
Causal Mechanism graph a new notation for capturing cause effect knowledge in software dependability
Reliability Engineering & System Safety, 2017Co-Authors: Fuqun Huang, Carol SmidtsAbstract:Abstract Understanding cause-effect relations between concepts in software dependability engineering is fundamental to various research or industrial activities. Cognitive maps are traditionally used to elicit and represent such knowledge; however they seem incapable of accurately representing complex Causal Mechanisms in dependability engineering. This paper proposes a new notation called Causal Mechanism Graph (CMG) to elicit and represent the cause-effect domain knowledge embedded in experts’ minds or described in the literature. CMG contains a new set of symbols elicited from domain experts to capture the recurring interaction Mechanisms between multiple concepts in software dependability engineering. Furthermore, compared to major existing graphic methods, CMG is particularly robust and suitable for mental knowledge elicitation: it allows one to represent the full range of cause-effect knowledge, accurately or fuzzily as one sees fit depending on the depth of knowledge he/she has. This feature combined with excellent reliability and validity poses CMG as a promising method that has the potential to be used in various areas, such as software dependability requirement elicitation, software dependability assessment and dependability risk control.
-
Causal Mechanism Graph ─ A new notation for capturing cause-effect knowledge in software dependability
Reliability Engineering & System Safety, 2017Co-Authors: Fuqun Huang, Carol SmidtsAbstract:Abstract Understanding cause-effect relations between concepts in software dependability engineering is fundamental to various research or industrial activities. Cognitive maps are traditionally used to elicit and represent such knowledge; however they seem incapable of accurately representing complex Causal Mechanisms in dependability engineering. This paper proposes a new notation called Causal Mechanism Graph (CMG) to elicit and represent the cause-effect domain knowledge embedded in experts’ minds or described in the literature. CMG contains a new set of symbols elicited from domain experts to capture the recurring interaction Mechanisms between multiple concepts in software dependability engineering. Furthermore, compared to major existing graphic methods, CMG is particularly robust and suitable for mental knowledge elicitation: it allows one to represent the full range of cause-effect knowledge, accurately or fuzzily as one sees fit depending on the depth of knowledge he/she has. This feature combined with excellent reliability and validity poses CMG as a promising method that has the potential to be used in various areas, such as software dependability requirement elicitation, software dependability assessment and dependability risk control.
Fuqun Huang - One of the best experts on this subject based on the ideXlab platform.
-
Causal Mechanism graph a new notation for capturing cause effect knowledge in software dependability
Reliability Engineering & System Safety, 2017Co-Authors: Fuqun Huang, Carol SmidtsAbstract:Abstract Understanding cause-effect relations between concepts in software dependability engineering is fundamental to various research or industrial activities. Cognitive maps are traditionally used to elicit and represent such knowledge; however they seem incapable of accurately representing complex Causal Mechanisms in dependability engineering. This paper proposes a new notation called Causal Mechanism Graph (CMG) to elicit and represent the cause-effect domain knowledge embedded in experts’ minds or described in the literature. CMG contains a new set of symbols elicited from domain experts to capture the recurring interaction Mechanisms between multiple concepts in software dependability engineering. Furthermore, compared to major existing graphic methods, CMG is particularly robust and suitable for mental knowledge elicitation: it allows one to represent the full range of cause-effect knowledge, accurately or fuzzily as one sees fit depending on the depth of knowledge he/she has. This feature combined with excellent reliability and validity poses CMG as a promising method that has the potential to be used in various areas, such as software dependability requirement elicitation, software dependability assessment and dependability risk control.
-
Causal Mechanism Graph ─ A new notation for capturing cause-effect knowledge in software dependability
Reliability Engineering & System Safety, 2017Co-Authors: Fuqun Huang, Carol SmidtsAbstract:Abstract Understanding cause-effect relations between concepts in software dependability engineering is fundamental to various research or industrial activities. Cognitive maps are traditionally used to elicit and represent such knowledge; however they seem incapable of accurately representing complex Causal Mechanisms in dependability engineering. This paper proposes a new notation called Causal Mechanism Graph (CMG) to elicit and represent the cause-effect domain knowledge embedded in experts’ minds or described in the literature. CMG contains a new set of symbols elicited from domain experts to capture the recurring interaction Mechanisms between multiple concepts in software dependability engineering. Furthermore, compared to major existing graphic methods, CMG is particularly robust and suitable for mental knowledge elicitation: it allows one to represent the full range of cause-effect knowledge, accurately or fuzzily as one sees fit depending on the depth of knowledge he/she has. This feature combined with excellent reliability and validity poses CMG as a promising method that has the potential to be used in various areas, such as software dependability requirement elicitation, software dependability assessment and dependability risk control.
Ruichu Cai - One of the best experts on this subject based on the ideXlab platform.
-
Causal Mechanism Transfer Network for Time Series Domain Adaptation in Mechanical Systems
arXiv: Learning, 2019Co-Authors: Ruichu Cai, Kok Soon Chai, Marianne Winslett, Xiaoyan Yang, Zhenjie ZhangAbstract:Data-driven models are becoming essential parts in modern mechanical systems, commonly used to capture the behavior of various equipment and varying environmental characteristics. Despite the advantages of these data-driven models on excellent adaptivity to high dynamics and aging equipment, they are usually hungry to massive labels over historical data, mostly contributed by human engineers at an extremely high cost. The label demand is now the major limiting factor to modeling accuracy, hindering the fulfillment of visions for applications. Fortunately, domain adaptation enhances the model generalization by utilizing the labelled source data as well as the unlabelled target data and then we can reuse the model on different domains. However, the mainstream domain adaptation methods cannot achieve ideal performance on time series data, because most of them focus on static samples and even the existing time series domain adaptation methods ignore the properties of time series data, such as temporal Causal Mechanism. In this paper, we assume that Causal Mechanism is invariant and present our Causal Mechanism Transfer Network(CMTN) for time series domain adaptation. By capturing and transferring the dynamic and temporal Causal Mechanism of multivariate time series data and alleviating the time lags and different value ranges among different machines, CMTN allows the data-driven models to exploit existing data and labels from similar systems, such that the resulting model on a new system is highly reliable even with very limited data. We report our empirical results and lessons learned from two real-world case studies, on chiller plant energy optimization and boiler fault detection, which outperforms the existing state-of-the-art method.
-
Causal discovery from discrete data using hidden compact representation
Neural Information Processing Systems, 2018Co-Authors: Ruichu Cai, Jie Qiao, Kun Zhang, Zhenjie Zhang, Zhifeng HaoAbstract:Causal discovery from a set of observations is one of the fundamental problems across several disciplines. For continuous variables, recently a number of Causal discovery methods have demonstrated their effectiveness in distinguishing the cause from effect by exploring certain properties of the conditional distribution, but Causal discovery on categorical data still remains to be a challenging problem, because it is generally not easy to find a compact description of the Causal Mechanism for the true Causal direction. In this paper we make an attempt to find a way to solve this problem by assuming a two-stage Causal process: the first stage maps the cause to a hidden variable of a lower cardinality, and the second stage generates the effect from the hidden representation. In this way, the Causal Mechanism admits a simple yet compact representation. We show that under this model, the Causal direction is identifiable under some weak conditions on the true Causal Mechanism. We also provide an effective solution to recover the above hidden compact representation within the likelihood framework. Empirical studies verify the effectiveness of the proposed approach on both synthetic and real-world data.
-
NeurIPS - Causal Discovery from Discrete Data using Hidden Compact Representation.
Advances in neural information processing systems, 2018Co-Authors: Ruichu Cai, Jie Qiao, Kun Zhang, Zhenjie Zhang, Zhifeng HaoAbstract:Causal discovery from a set of observations is one of the fundamental problems across several disciplines. For continuous variables, recently a number of Causal discovery methods have demonstrated their effectiveness in distinguishing the cause from effect by exploring certain properties of the conditional distribution, but Causal discovery on categorical data still remains to be a challenging problem, because it is generally not easy to find a compact description of the Causal Mechanism for the true Causal direction. In this paper we make an attempt to find a way to solve this problem by assuming a two-stage Causal process: the first stage maps the cause to a hidden variable of a lower cardinality, and the second stage generates the effect from the hidden representation. In this way, the Causal Mechanism admits a simple yet compact representation. We show that under this model, the Causal direction is identifiable under some weak conditions on the true Causal Mechanism. We also provide an effective solution to recover the above hidden compact representation within the likelihood framework. Empirical studies verify the effectiveness of the proposed approach on both synthetic and real-world data.
Zhenjie Zhang - One of the best experts on this subject based on the ideXlab platform.
-
Causal Mechanism Transfer Network for Time Series Domain Adaptation in Mechanical Systems
arXiv: Learning, 2019Co-Authors: Ruichu Cai, Kok Soon Chai, Marianne Winslett, Xiaoyan Yang, Zhenjie ZhangAbstract:Data-driven models are becoming essential parts in modern mechanical systems, commonly used to capture the behavior of various equipment and varying environmental characteristics. Despite the advantages of these data-driven models on excellent adaptivity to high dynamics and aging equipment, they are usually hungry to massive labels over historical data, mostly contributed by human engineers at an extremely high cost. The label demand is now the major limiting factor to modeling accuracy, hindering the fulfillment of visions for applications. Fortunately, domain adaptation enhances the model generalization by utilizing the labelled source data as well as the unlabelled target data and then we can reuse the model on different domains. However, the mainstream domain adaptation methods cannot achieve ideal performance on time series data, because most of them focus on static samples and even the existing time series domain adaptation methods ignore the properties of time series data, such as temporal Causal Mechanism. In this paper, we assume that Causal Mechanism is invariant and present our Causal Mechanism Transfer Network(CMTN) for time series domain adaptation. By capturing and transferring the dynamic and temporal Causal Mechanism of multivariate time series data and alleviating the time lags and different value ranges among different machines, CMTN allows the data-driven models to exploit existing data and labels from similar systems, such that the resulting model on a new system is highly reliable even with very limited data. We report our empirical results and lessons learned from two real-world case studies, on chiller plant energy optimization and boiler fault detection, which outperforms the existing state-of-the-art method.
-
Causal discovery from discrete data using hidden compact representation
Neural Information Processing Systems, 2018Co-Authors: Ruichu Cai, Jie Qiao, Kun Zhang, Zhenjie Zhang, Zhifeng HaoAbstract:Causal discovery from a set of observations is one of the fundamental problems across several disciplines. For continuous variables, recently a number of Causal discovery methods have demonstrated their effectiveness in distinguishing the cause from effect by exploring certain properties of the conditional distribution, but Causal discovery on categorical data still remains to be a challenging problem, because it is generally not easy to find a compact description of the Causal Mechanism for the true Causal direction. In this paper we make an attempt to find a way to solve this problem by assuming a two-stage Causal process: the first stage maps the cause to a hidden variable of a lower cardinality, and the second stage generates the effect from the hidden representation. In this way, the Causal Mechanism admits a simple yet compact representation. We show that under this model, the Causal direction is identifiable under some weak conditions on the true Causal Mechanism. We also provide an effective solution to recover the above hidden compact representation within the likelihood framework. Empirical studies verify the effectiveness of the proposed approach on both synthetic and real-world data.
-
NeurIPS - Causal Discovery from Discrete Data using Hidden Compact Representation.
Advances in neural information processing systems, 2018Co-Authors: Ruichu Cai, Jie Qiao, Kun Zhang, Zhenjie Zhang, Zhifeng HaoAbstract:Causal discovery from a set of observations is one of the fundamental problems across several disciplines. For continuous variables, recently a number of Causal discovery methods have demonstrated their effectiveness in distinguishing the cause from effect by exploring certain properties of the conditional distribution, but Causal discovery on categorical data still remains to be a challenging problem, because it is generally not easy to find a compact description of the Causal Mechanism for the true Causal direction. In this paper we make an attempt to find a way to solve this problem by assuming a two-stage Causal process: the first stage maps the cause to a hidden variable of a lower cardinality, and the second stage generates the effect from the hidden representation. In this way, the Causal Mechanism admits a simple yet compact representation. We show that under this model, the Causal direction is identifiable under some weak conditions on the true Causal Mechanism. We also provide an effective solution to recover the above hidden compact representation within the likelihood framework. Empirical studies verify the effectiveness of the proposed approach on both synthetic and real-world data.
Tatsiana Aneichyk - One of the best experts on this subject based on the ideXlab platform.
-
dissecting the Causal Mechanism of x linked dystonia parkinsonism by integrating genome and transcriptome assembly
Cell, 2018Co-Authors: Tatsiana Aneichyk, William T. Hendriks, Rachita Yadav, Christine A. Vaine, R. Collins, David Shin, Aloysius DomingoAbstract:Summary X-linked Dystonia-Parkinsonism (XDP) is a Mendelian neurodegenerative disease that is endemic to the Philippines and is associated with a founder haplotype. We integrated multiple genome and transcriptome assembly technologies to narrow the Causal mutation to the TAF1 locus, which included a SINE-VNTR-Alu (SVA) retrotransposition into intron 32 of the gene. Transcriptome analyses identified decreased expression of the canonical cTAF1 transcript among XDP probands, and de novo assembly across multiple pluripotent stem-cell-derived neuronal lineages discovered aberrant TAF1 transcription that involved alternative splicing and intron retention (IR) in proximity to the SVA that was anti-correlated with overall TAF1 expression. CRISPR/Cas9 excision of the SVA rescued this XDP-specific transcriptional signature and normalized TAF1 expression in probands. These data suggest an SVA-mediated aberrant transcriptional Mechanism associated with XDP and may provide a roadmap for layered technologies and integrated assembly-based analyses for other unsolved Mendelian disorders.
-
Dissecting the Causal Mechanism of X-Linked Dystonia-Parkinsonism by Integrating Genome and Transcriptome Assembly
bioRxiv, 2017Co-Authors: Tatsiana Aneichyk, William T. Hendriks, Rachita Yadav, Christine A. Vaine, Alexei Stortchevoi, Benjamin B. Currall, Ryan L. Collins, David Shin, Harrison BrandAbstract:X-linked Dystonia-Parkinsonism (XDP) is a Mendelian neurodegenerative disease endemic to the Philippines. We integrated genome and transcriptome assembly with induced pluripotent stem cell-based modeling to identify the XDP Causal locus and potential pathogenic Mechanism. Genome sequencing identified novel variation that was shared by all probands and three recombination events that narrowed the Causal locus to a genomic segment including TAF1. Transcriptome assembly in neural derivative cells discovered novel TAF1 transcripts, including a truncated transcript exclusively observed in probands that involved aberrant splicing and intron retention (IR) associated with a SINE-VNTR-Alu (SVA)-type retrotransposon insertion. This IR correlated with decreased expression of the predominant TAF1 transcript and altered expression of neurodevelopmental genes; both the IR and aberrant TAF1 expression patterns were rescued by CRISPR/Cas9 excision of the SVA. These data suggest a unique genomic cause of XDP and may provide a roadmap for integrative genomic studies in other unsolved Mendelian disorders.