We introduce Joint Causal Inference (JCI), a powerful formulation of causal discovery from multiple datasets that allows to jointly learn both the causal structure and targets of interventions from statistical independences in pooled data. Compared with existing constraint-based approaches for causal discovery from multiple data sets, JCI offers several advantages: it allows for several differe...

Earlier studies in 1980s found no causal links between IT investments and productivity, since then a growing body of research has been investigating such links at a much finer-level of analysis. Yet the results have been inconclusive. We attribute the mixed findings to an incomplete causal chain analysis, specifically the exclusion of key constructs such as strategic alignment which would allow...

Discrimination discovery is to unveil discrimination against a specific individual by analyzing the historical dataset. In this paper, we develop a general technique to capture discrimination based on the legally grounded situation testing methodology. For any individual, we find pairs of tuples from the dataset with similar characteristics apart from belonging or not to the protected-by-law gr...

In this paper we address the issue of causal rhythmic analysis, primarily towards predicting the locations of musical beats such that they are consistent with a musical audio input. This will be a key component required for a system capable of automatic accompaniment with a live musician. We are implementing our approach as part of the aubio real-time audio library. While performance for this c...

National IQs assessed by the Progressive Matrices were calculated for 60 nations and examined in relation to per capita incomes in the late 1990s and to post World War Two rates of economic growth. It was found that national IQs are correlated at 0.757 with real GDP (Gross Domestic Product) per capita 1998 and 0.706 with per capita GNP (Gross National Product) 1998; and at 0.605 with the growth...

We present a symbolic machinery that admits both probabilistic and causal information about a given domain and produces probabilistic statements about he effect of actions and the impact of observations. Thecalculus admits two types of conditioning operators: ordinary Bayes conditioning, P(y]X = z), which represents he observation X z, and causal conditioning, P(yldo(X = x)), read the probabili...

Recent advances in classical planning have used the SAS+ formalism, and several effective heuristics have been developed based on the SAS+ formalism. Comparing to the traditional STRIPS/ADL formalism, SAS+ is capable of capturing vital information such as domain transition structures and causal dependencies. In this paper, we propose a new SAS+ based incomplete planning approach. Instead of usi...

We address the problem of estimating the effect of intervening on a set of variables X from experiments on a different set, Z, that is more accessible to manipulation. This problem, which we call z-identifiability, reduces to ordinary identifiability when Z = ∅ and, like the latter, can be given syntactic characterization using the do-calculus [Pearl, 1995; 2000]. We provide a graphical necessa...