## Causal Discovery Using Adaptive Logics. Towards a more realistic heuristics for human causal learning (2004)

Citations: | 1 - 0 self |

### BibTeX

@MISC{Dyck04causaldiscovery,

author = {Maarten Van Dyck},

title = {Causal Discovery Using Adaptive Logics. Towards a more realistic heuristics for human causal learning},

year = {2004}

}

### OpenURL

### Abstract

### Citations

449 | A treatise of Human Nature - Hume - 1739 |

204 | The Direction of Time - Reichenbach |

200 |
Scientific Explanation and the Causal Structure of the World
- Salmon
- 1984
(Show Context)
Citation Context ...ut this implication, let us briefly try to understand what this condition comes down to. It is widely recognized to be a generalization of Reichenbach’s common cause condition (on this condition, see =-=[13, 15]-=-); and 6sthis condition was meant to capture two ideas: that all correlations have a causal explanation, and that common causes screen off their correlated effects. If the causal Markov condition hold... |

158 | A theory of causal learning in children: causal maps and Bayes nets
- Gopnik, Glymore, et al.
- 2004
(Show Context)
Citation Context ...ships in ways that can be described as causal Bayes nets, and moreover that even children might make causal inferences without discriminating between potential causes and potential effects beforehand =-=[8]-=-. 23sThe heuristics proposed in the present article is clearly based on the assumptions underlying the Bayes nets approach, and its final output will always be a Bayes net (or a disjunction of Bayes n... |

89 |
Inferring causal networks from observations and interventions
- Steyvers, Tenenbaum, et al.
(Show Context)
Citation Context ... also play a useful heuristic role in this, by pointing towards interesting hypotheses to be investigated. 20 This guessing could be — partly — corrected by adopting Bayesian learning algorithms (cf. =-=[17]-=-) 24sover, in [8] it is exactly this problem that is singled out as one of the major obstacles for the causal Bayes net approach to provide realistic models of human reasoning. A R1, R2, R3 and d–sepa... |

79 | The mind’s arrows: Bayes nets and graphical causal models in psychology - Glymour - 2001 |

60 |
2000), “A Survey of Inconsistency-Adaptive Logics”, in: Batens, Diderik et. al. (Eds.) Frontiers of Paraconsistent Logic
- Batens
(Show Context)
Citation Context ...hind adaptive logics. 3 Adaptive logics Adaptive logics were originally developed to deal with a particular problem arising when a set of premisses give rise to an inconsistency (for an overview, see =-=[2]-=-). It is of course well known that classical logic turns out to be completely impotent in this situation: the consequence set is trivial, following the classical property of ex falso quodlibet. In vie... |

25 |
An overview of the representation and discovery of causal relationships using Bayesian networks
- Cooper
- 1999
(Show Context)
Citation Context ...likeliness has to be interpreted, and hence about how justified this assumption is; I will not comment on this issue, but refer the interested reader to [14]. 12 I follow the presentation as given in =-=[4]-=-. 11sthat X is d–separated from Y given Z, then in the probability distribution P we have that X and Y are conditionally dependent given Z, where X and Y are not empty but Z may be. Judea Pearl opts f... |

22 |
Humphreys: ‘Are there algorithms that discover causal structure?’, Synthese 121
- Freedman, Paul
(Show Context)
Citation Context ...ification of adding the present article to the already existing literature on causal discovery. Section 3 is not supposed to contain anything new for readers al1 For a very critical appreciation, see =-=[5]-=-. 2 See the concluding section, however, for an important caveat with respect to this claim. There are two possible reasons for claiming the algorithms to be unlikely psychological models. The reformu... |

20 |
Uniform consistency in causal inference
- Robins, Scheines, et al.
- 2003
(Show Context)
Citation Context ... 11 There is some controversy about how this unlikeliness has to be interpreted, and hence about how justified this assumption is; I will not comment on this issue, but refer the interested reader to =-=[14]-=-. 12 I follow the presentation as given in [4]. 11sthat X is d–separated from Y given Z, then in the probability distribution P we have that X and Y are conditionally dependent given Z, where X and Y ... |

15 | Causal maps and Bayes nets: a cognitive and computational account of theory-formation - Gopnik, Glymour - 2002 |

14 |
Independence, invariance, and the causal Markov condition
- Hausman, Woodward
- 1999
(Show Context)
Citation Context ...r this to be a justifiable methodology a justification for the Markov condition has to be provided. I will not attempt such here, but refer the reader to the excellent article by Hausman and Woodward =-=[9]-=-. Let me suffice with a few remarks on this topic. First consider the case of mediated causal influences. It is clear that many of the ideas associated with causal mechanisms would fail if indirect ca... |

13 | On a logic of induction
- Batens
- 2005
(Show Context)
Citation Context ...an permitted by classical logic, where the particular criteria for marking (and definitions of abnormality) will determine when “further” means “too far”. Examples are an adaptive logic for induction =-=[1]-=-, and one for abduction [11]. As was already stressed extensively by David Hume, causal reasoning is a species of inductive reasoning, so it will not come as a surprise that the adaptive logic for cau... |

11 | Ampliative adaptive logics and the foundation of logic-based approaches to abduction
- Meheus, Verhoeven, et al.
(Show Context)
Citation Context ...ogic, where the particular criteria for marking (and definitions of abnormality) will determine when “further” means “too far”. Examples are an adaptive logic for induction [1], and one for abduction =-=[11]-=-. As was already stressed extensively by David Hume, causal reasoning is a species of inductive reasoning, so it will not come as a surprise that the adaptive logic for causal discovery will also be o... |

3 |
Causal diversity and the markov condition
- Cartwright
- 1999
(Show Context)
Citation Context ...irect to the direct causes) will not have any influence on X. The case of correlated effects of common causes is similar, but more controversial (the main critiques coming from Nancy Cartwright, e.g. =-=[3]-=-). 9 There seems to be something very Markov condition, since this only says that the set of parents screen off from non-effects; it is silent about any other set of variables. 9 The quantum-mechanica... |