The Book of Why: The New Science of Cause and Effect
By Mélanie Frappier. 2018
2,456 citations recorded by the source corpus.
This week on the Science podcast, Judea Pearl and Dana Mackenzie discuss strategies for causal thinking and describe its implications for artificial intelligence.
Subfields: Management Information Systems, Causal Inference & Discovery
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Management Information Systems
- Yogesh Kumar Dwivedi
- Robert R. Hoffman
- Nripendra P. Rana
- William S. Isaac
- Kaihua Tang
- Markku V. J. Maula
- Fei Sun
- Sharad Goel
In Causal Inference & Discovery
- Bernhard Schölkopf
- Jan Peters
- Dominik Janzing
- Kun Zhang
- Joris M. Mooij
- Stefan Bauer
- Yoshua Bengio
- Anirudh Goyal
Citations in the corpus
- Amnesic Probing: Behavioral Explanation with Amnesic Counterfactuals (2020, background): “This approach allows us to ask the counterfactual question: “How will the prediction of a task differ without access to some property?” (Pearl and Mackenzie, 2018).”
- Abductive Commonsense Reasoning (2019, neutral): “…has long been considered to be at the core of understanding narratives Hobbs et al. (1988), reading between the lines Norvig (1987); Charniak & Shimony (1990), reasoning about everyday situations Peirce (1965b); Andersen (1973), and counterfactual reasoning Pearl (2002); Pearl & Mackenzie (2018).”
- Explanation in Human-AI Systems: A Literature Meta-Review, Synopsis of Key Ideas and Publications, and Bibliography for Explainable AI (2019, background): “…to, and perhaps even the same as causal reasoning, since explanations often refer to causation or causal mechanisms, and causal analyses are believed to have explanatory value (Cohen, 2008; Halpern and Pearl, 2005a,b; Krull and Anderson, 1997; Pearl, 1988; Pearl and Mackenzie, 2018; Trout, 2002).”
- Explanation in Artificial Intelligence: Insights from the Social Sciences (2017, background): “In Table 3, I propose a simple model for explanatory questions based on Pearl and Mackenzie’s Ladder of Causation [141].”
- Simpson's Paradox in COVID-19 Case Fatality Rates: A Mediation Analysis of Age-Related Causal Effects (2020, background): “4Pearl and Mackenzie[20] give the illustrative example of a drug (treatment) that works by activating some proteins (mediator) inside the body before jointly attacking the disease: the drug is useless without the activated proteins (so the direct effect is zero) and the activated protein is useless ”
- Learning Neural Causal Models from Unknown Interventions (2019, background): “81 Â 1 0 1 0 0 0 0 0 0 1-hot sample B = [1, 0] 0.”
- Learning First-Order Symbolic Representations for Planning from the Structure of the State Space (2019, methodology): “Using Pearl’s terminology [33], the input graphs can be regarded as defining the complete space of possible causal interventions that allow us to recover the causal structure of the domain in a first-order language.”
- Woulda, Coulda, Shoulda: Counterfactually-Guided Policy Search (2018, background): “This example tries to illustrate the everyday human capacity to reason about alternate, counterfactual outcomes of past experience with the goal of “mining worlds that could have been” (Pearl & Mackenzie, 2018).”
- Participatory Problem Formulation for Fairer Machine Learning Through Community Based System Dynamics (2020, background): “Psychological research has shown that human causal inference is based on a priori intuitive theories about the causal structure of the problem to be intervened on (Tenenbaum & Griffiths, 2003; Pearl & Mackenzie, 2018).”
- Unbiased Scene Graph Generation From Biased Training (2020, neutral): “More detailed background knowledge can be found in [40, 41, 57].”
- COPHY: Counterfactual Learning of Physical Dynamics (2019, background): “(Pearl, 2009; Pearl & McKenzie, 2018) for dealing with causal inference (Spirtes, 2010).”
- Meaningful Explanations of Black Box AI Decision Systems (2019, methodology): “…inference and learning has developed tools, such as Causal Graphical Models and Structural Causal Models (Peters, Janzing, and Scholkopf 2017; Pearl and MacKenzie 2018; Bareinboim and Pearl 2016), to answer certain interventional questions (what if some-thing changes?) or retrospective…”
- Two Causal Principles for Improving Visual Dialog (2019, methodology): “The technique is do -calculus introduced in [28, 29].”
- Stakeholders in Explainable AI (2018, neutral): “Others have used the term legibility (Kirsch 2017) while recent thinking once again emphasises ‘strong’ notions of explainability in causal terms (Pearl and Mackenzie 2018).”
- Machine Learning, Health Disparities, and Causal Reasoning (2018, background): “Why is this being brought to the fore only now, in the context of machine learning and social inequity?”
- Artificial Intelligence (AI): Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy (2019, background): “Under what circumstances will AI be su ffi cient and when will it need to be supported by causal reasoning or simulation (Pearl & MacKenzie, 2018)?”