├── CHANGELOG ├── CITEME.md ├── CONTRIBUTORS.md ├── README.md ├── VERSION ├── assets └── pearl_penrose.jpg ├── contribution_guide.md ├── looper.md ├── nuggets ├── LN1.md └── looper_nuggets.md └── src └── README.md /CHANGELOG: -------------------------------------------------------------------------------- 1 | 0.3.1 2 | * New Sections: 3 | * Thesis: Causal Representation learning and point process. 4 | * Review: Imbens's review. 5 | * Conference update: LLMs are causal parrots. 6 | * ML papers: Causal bandits, causal threatment outcome prediction. 7 | 8 | 0.3.0 9 | * New books: Chernozhukov et. al. (2024), Hurwitz-Thompson (2023). 10 | * Logo: Penrose meets Pearl. 11 | * Nobel Memorial Prize 2021, Laureates names. 12 | * Datasets Hunninton-Klein R package. 13 | * Software section clean-up: Sub-sections. 14 | * Additional conference links. 15 | * Simpson's paradox section updates. 16 | * MOOC; mixsessions.io Course material. -------------------------------------------------------------------------------- /CITEME.md: -------------------------------------------------------------------------------- 1 | ``` 2 | @misc{suezen2018a, 3 | author = {Mehmet S{\"u}zen et. al.}, 4 | title = {A resource list for causality in statistics, data science and physics}, 5 | year = {2018-2023}, 6 | publisher = {GitHub}, 7 | journal = {GitHub repository}, 8 | howpublished = {\url{https://github.com/msuzen/looper}}, 9 | } 10 | ``` 11 | -------------------------------------------------------------------------------- /CONTRIBUTORS.md: -------------------------------------------------------------------------------- 1 | 2 | * Mehmet Suzen (Alumni Frankfurt) : Principal Author & Editor 3 | 4 | # Contributors 5 | 6 | * Uri Shalit (Technion). 7 | * Rodney Beard (Glasgow). 8 | * Nino Malekovic (The Hague/Basel). 9 | * Cho Dong-Hwan github:dhjo70. 10 | * Jeff Willingham (Denver). 11 | * Professor Colin Cameron (UC Davis). 12 | * James T. Metz (Chicago). 13 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # looper : A resource list for causality in statistics, data science and physics 2 | 3 |

4 | 5 | Penrose meets Pearl, (c) 2021 6 |

7 | 8 | 9 | Our honour to be mentioned by Judea Pearl on [twitter](https://twitter.com/ceobillionaire/status/1388630546797023232). 10 | 11 | ``` 12 | 13 | Aspiration to learn everything from data alone has 14 | kept the ML community away from science. 15 | 16 | Judea Pearl 17 | 18 | ``` 19 | 20 | A resource list, code snippets or small scale software solutions for causal analysis in different areas. The name is inspired from the movie [looper](https://en.wikipedia.org/wiki/Looper_(film)), which has a premise of time-like loops, probably the most complex causal subject from physics point of view. 21 | 22 | The main entry is a markdown file as follows, any looper specific internal examples are lined there too : 23 | 24 | * [A resource list for causality in statistics, data science and physics](looper.md) 25 | Other fields of course such as econometrics, epidemiology and many more. 26 | 27 | ## Looper Nuggets 28 | 29 | Looper Nuggets mimick a glossary of terms and concepts in causal inference, though they are entry to 30 | understanding concepts in pedagogical manner. See the list of 31 | them [here](nuggets/looper_nuggets.md). 32 | 33 | ## License 34 | 35 | This repository and all contributions are licensed under 36 | [![License: CC BY 4.0](https://i.creativecommons.org/l/by/4.0/88x31.png)](https://creativecommons.org/licenses/by/4.0/) 37 | 38 | ## Citing the repo 39 | 40 | Please attribute this work as follows 41 | 42 | ``` 43 | @misc{suezen2018a, 44 | author = {Mehmet S{\"u}zen et. al.}, 45 | title = {A resource list for causality in statistics, data science and physics}, 46 | year = {2018}, 47 | publisher = {GitHub}, 48 | journal = {GitHub repository}, 49 | howpublished = {\url{https://github.com/msuzen/looper}}, 50 | } 51 | ``` 52 | If you are embedding specific release, use the version link, for example for `https://github.com/msuzen/looper/tree/v0.1.2` 53 | in `howpublished` tag. 54 | 55 | ## Contributions 56 | 57 | Please send a pull request or create an issue for suggestions or codes. See [Basic how to contribute guide](contribution_guide.md) 58 | 59 | -------------------------------------------------------------------------------- /VERSION: -------------------------------------------------------------------------------- 1 | 0.3.1 -------------------------------------------------------------------------------- /assets/pearl_penrose.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/msuzen/looper/be3d5d910ea8db422602f3e398cd545706bf74d9/assets/pearl_penrose.jpg -------------------------------------------------------------------------------- /contribution_guide.md: -------------------------------------------------------------------------------- 1 | # How to contribute to looper 2 | 3 | The looper is aim at providing a resource list, code snippets or small scale software solutions 4 | for causal analysis in different areas. You can contribute to looper by sending a pull request 5 | and you would automatically accept that your contributions might be accepted and absorb into 6 | looper repository under our open licencse, CC-BY 4.0. 7 | 8 | ## Adding a resource link 9 | 10 | * Find an appropriate section in the file `looper.md`. 11 | * There is no standard formatting except placing the title 12 | and information on the resource location and a link. 13 | * For readibility, use a convention of placing two spaces to have line breaks. 14 | See [How to insert a line breakin markdown](https://stackoverflow.com/questions/26626256/how-to-insert-a-line-break-br-in-markdown) 15 | * The sections of books and papers are in reverse-chronological order. 16 | * For software links, try to procide as much information as possible. 17 | 18 | ## Adding software solution or code 19 | 20 | * Use `src` directory and place your solution under this with a unique name. 21 | * Place an internal link to your directory in software section `looper.md`. 22 | 23 | ## Pull request 24 | 25 | * You would need to branch/fork from master and add your contributions. 26 | * Add your name and possibily affiliation in `CONTRIBUTORS` file if you are not there yet. 27 | 28 | ## Contribute via issue 29 | 30 | You can also contribute via creating an issue if using github is too time consuming for you. 31 | -------------------------------------------------------------------------------- /looper.md: -------------------------------------------------------------------------------- 1 | # [A resource list for causality in statistics, data science and physics](https://github.com/msuzen/looper) 2 | 3 | ##### Table of Contents 4 | [Editor's Selection](#editors-selection). 5 | [Books](#books) 6 | [Papers general](#papers-general) 7 | [Causal Discovery](#causal-discovery) 8 | [Simpson's Paradox](#simpsons-paradox) 9 | [Machine Learning](#machine-learning) 10 | [Fairness](#fairness) 11 | [Physics](#physics) 12 | [Thesis](#thesis) 13 | [Reviews](#review) 14 | [Software](#software) 15 | [Datasets](#datasets) 16 | [MOOCs](#moocs) 17 | [Blog Posts](#blog-posts) 18 | [Quotes](#quotes) 19 | [Video Lectures](#video-lectures). 20 | [Academics](#academics). 21 | [Communities](#communities-conferences) 22 | 23 | ## Editors Selection 24 | Repeating from other sections, Highly important 25 | Order from beginner to advanced. 26 | 27 | * Judea Pearl and Mackenzie. 28 | The Book of Why: The New Science of Cause and Effect (2018). 29 | [amzn](https://www.amzn.com/dp/046509760X) | [Pearl's additional links](http://bayes.cs.ucla.edu/WHY/) 30 | 31 | * The Seven Tools of Causal Inference, with Reflections on Machine Learning. 32 | Judea Pearl. 33 | Communications of the ACM, March 2019, Vol. 62 No. 3, Pages 54-60. 34 | [url](https://cacm.acm.org/magazines/2019/3/234929-the-seven-tools-of-causal-inference-with-reflections-on-machine-learning/fulltext) 35 | [pdf-reprint](https://ftp.cs.ucla.edu/pub/stat_ser/r481-reprint.pdf). 36 | [interview](https://www.youtube.com/watch?v=CsMV5o3hotY). 37 | 38 | * Causal Inference: History, Perspectives, Adventures, and Unification (An Interview with Judea Pearl). 39 | Observational Studies, 8, 2, (2022), p1-14. 40 | [url](https://muse.jhu.edu/article/867087) | [pdf](https://muse.jhu.edu/pub/56/article/867087/pdf). 41 | 42 | * Pearl, Glymour and Jewell. 43 | Causal Inference in Statistics: A Primer (2016). 44 | [amzn](https://www.amzn.com/dp/1119186846). 45 | [Ch4-pdf](http://web.cs.ucla.edu/~kaoru/primer-ch4.pdf). 46 | [tweet-solution-manual](https://twitter.com/yudapearl/status/1484023795811696642). 47 | Self-study by Bruno Goncalves [github](https://github.com/DataForScience/Causality) 48 | 49 | * Nobel Memorial Economics Prize 2021 on causal discovery, scientific summary. 50 | Answering Causal Questions Using Observational Data. (2021) 51 | (David Card, Joshua Angrist, and Guido Imbens) 52 | [pdf](https://www.nobelprize.org/uploads/2021/10/advanced-economicsciencesprize2021.pdf). 53 | 54 | * Causality, determinism, and physics. 55 | Julio Geo-Banacloche. 56 | American Journal of Physics 90, 809 (2022); [doi](https://doi.org/10.1119/5.0087017). 57 | 58 | * Causal diagrams for empirical research. 59 | Judea Pearl (1995). 60 | [jstor](https://www.jstor.org/stable/2337329) | [pdf-UCLA](http://bayes.cs.ucla.edu/R218-B.pdf). 61 | `Reasoning on Graphs: d-seperation, back/front-door` 62 | 63 | * Paul W. Holland. 64 | Statistics and Causal Inference. 65 | Journal of the American Statistical Association. 66 | Dec., 1986, Vol. 81, No. 396 (Dec., 1986), pp. 945-960 [jstor](https://www.jstor.org/stable/2289064). | [pdf-from-David-Bleie](https://www.cs.columbia.edu/~blei/fogm/2020F/readings/Holland1986.pdf) 67 | `Paper that coins the term: Fundamental Problem of Causal Inference` 68 | `It means there is no way to observe causality on the single entity simultaneously.` 69 | 70 | * Probabilistic and Causal Inference: The Works of Judea Pearl. (2022) 71 | Editors: Hector Geffner,Rina Dechter,Joseph Y. Halpern. 72 | ACM Books [doi](https://dl.acm.org/doi/book/10.1145/3501714). 73 | `An excellent technical reviews of Pearlian approach to causal inference` 74 | 75 | ## Books 76 | 77 | Reverse chronological order, both technical and popular. 78 | 79 | * Applied Causal Inference Powered by ML and AI 80 | V. Chernozhukov, C. Hansen, N. Kallus, M. Spindler, V. Syrgkanis (2024) 81 | [offfical-web](https://causalml-book.org) | [pdf](https://causalml-book.org/assets/chapters/CausalML_book.pdf) 82 | 83 | * A Mathematical Introduction to Causality 84 | Lecture Notes (2023) 85 | Patrick Forré and Joris M. Mooij 86 | [UvA's-pdf](https://staff.fnwi.uva.nl/j.m.mooij/articles/causality_lecture_notes_2023.pdf) 87 | 88 | * A First Course in Causal Inference. 89 | Peng Ding. 90 | [arXiv](https://arxiv.org/abs/2305.18793) (2023) 91 | 92 | * Applied Causal Inference 93 | Uday Kamath, Kenneth Graham, Mitchell Naylor (2023) 94 | [online-version](https://appliedcausalinference.github.io) | [lean-pub](https://leanpub.com/appliedcausalinference) 95 | 96 | * Causal Artificial Intelligence: The Next Step in Effective Business AI 97 | Judith Hurwitz, John Thompson (2023) 98 | [amz](https://amzn.com/dp/1394184131) 99 | 100 | * Causal Inference & Discovery in Python. (2023) 101 | From Machine Learning & Pearlian Perspective. 102 | Aleksander Molak. 103 | [causalpython](https://causalpython.io). | [amz](https://amzn.com/dp/1804612987) 104 | 105 | * Causal Inference for Data Science. 106 | Aleix Ruiz de Villa (2023) 107 | [manning](https://www.manning.com/books/causal-inference-for-data-science). 108 | 109 | * Causal Analysis. 110 | Impact Evaluation and Causal Machine Learning with Applications in R. 111 | Martin Huber (2023). 112 | MIT Press [url](https://mitpress.mit.edu/9780262545914/causal-analysis/) | R package [causalweight](https://cran.r-project.org/web/packages/causalweight/index.html) | [R examples](https://www.unifr.ch/appecon/en/assets/public/uploads/causal%20analysis%20-%20R%20examples.txt) | [academic page](https://www.unifr.ch/appecon/en/research/text-book-causal-analysis.html) 113 | 114 | * Causal Inference in Python. 115 | Matheus Facure. 116 | [Oreilly](https://www.oreilly.com/library/view/causal-inference-in/9781098140243/) (2023). 117 | 118 | * Causal Inference for The Brave and True. (2022). 119 | Matheus Facure. 120 | [e-book](https://matheusfacure.github.io/python-causality-handbook/landing-page.html) | [github](https://github.com/matheusfacure/python-causality-handbook) 121 | 122 | * Probabilistic and Causal Inference: The Works of Judea Pearl. (2022) 123 | Editors: Hector Geffner,Rina Dechter,Joseph Y. Halpern. 124 | ACM Books [doi](https://dl.acm.org/doi/book/10.1145/3501714). 125 | 126 | * Microeconometrics using Stata: Volume 2: Nonlinear Models and Causal Inference Methods. (2022). 127 | Cameron, A. C., and P. K. Trivedi. 128 | Stata Press [url](https://www.routledge.com/Microeconometrics-Using-Stata-Second-Edition-Volume-II-Nonlinear-Models/Cameron-Trivedi/p/book/9781597183628) 129 | 130 | * Nick Huntington-Klein. 131 | The Effect: An Introduction to Research Design and Causality (2021). 132 | [bookdown](https://theeffectbook.net/index.html) | [amzn](https://amzn.com/dp/1032125780). 133 | [code-supplement](https://github.com/NickCH-K/causalbook) | [datasets-R](https://cran.r-project.org/web/packages/causaldata/index.html). 134 | 135 | * Causal Machine Learning. 136 | Robert Ness (2022-). 137 | [in-progress](https://www.manning.com/books/causal-machine-learning) | [lecture notes](https://bookdown.org/robertness/causalml/docs/) 138 | 139 | * Causal Inference, The Mixtape. 140 | Scott Cunningham. 141 | Yale University Press [url](https://yalebooks.yale.edu/book/9780300251685/causal-inference/) | [online-version](https://mixtape.scunning.com) (2021) 142 | 143 | * Hernán & Robins. 144 | Causal Inference: What If (2020). 145 | CRC Press. 146 | [online](http://bit.ly/2mSeeXI) | [pdf](https://cdn1.sph.harvard.edu/wp-content/uploads/sites/1268/2021/01/ciwhatif_hernanrobins_31dec20.pdf). 147 | [python supplement](https://github.com/jrfiedler/causal_inference_python_code) 148 | 149 | * Brady Neal. 150 | Introduction to Causal Inference from a Machine Learning Perspective. (2020). 151 | [pdf](https://www.bradyneal.com/Introduction_to_Causal_Inference-Dec17_2020-Neal.pdf) 152 | 153 | * Kohavi, Ron, Diane Tang, and Ya Xu. 154 | Trustworthy Online Controlled Experiments: A Practical Guide to A/B Testing. 155 | Cambridge University Press. [www](https://experimentguide.com/). (2020). 156 | 157 | * Experimental Design : Causal Inference 158 | Stefan Wager. 159 | Class notes [pdf](https://statweb.stanford.edu/~owen/courses/363/stats361.pdf). 160 | [url](https://statweb.stanford.edu/~owen/courses/363/). 161 | 162 | * Pearl and Mackenzie. 163 | The Book of Why: The New Science of Cause and Effect (2018). 164 | [amzn](https://www.amzn.com/dp/046509760X) | [Pearl's additional links](http://bayes.cs.ucla.edu/WHY/) 165 | 166 | * Rosenbaum. 167 | Observation and Experiment: An Introduction to Causal Inference (2017). 168 | [amzn](https://www.amzn.com/dp/067497557X/). 169 | 170 | * Jonas Peters, Dominik Janzing and Bernhard Schoelkopf. 171 | Elements of Causal Inference: Foundations and Learning Algorithms (2017). 172 | [mitpress](https://mitpress.mit.edu/books/elements-causal-inference) | [pdf](https://library.oapen.org/bitstream/handle/20.500.12657/26040/11283.pdf). 173 | 174 | * Pearl, Glymour and Jewell. 175 | Causal Inference in Statistics: A Primer (2016). 176 | [amzn](https://www.amzn.com/dp/1119186846). 177 | [Ch4-pdf](http://web.cs.ucla.edu/~kaoru/primer-ch4.pdf). 178 | [tweet-solution-manual](https://twitter.com/yudapearl/status/1484023795811696642). 179 | Self-study by Bruno Goncalves [github](https://github.com/DataForScience/Causality) 180 | 181 | * Mastering 'Metrics: The Path from Cause to Effect. (2015). 182 | Angrist, J.D. and J.-S.Pischke. 183 | Princeton University Press [url](https://press.princeton.edu/books/paperback/9780691152844/mastering-metrics) 184 | 185 | * Morgan & Winship. 186 | Counterfactuals and Causal Inference (2nd edition) (2015). 187 | [amzn](https://www.amzn.com/dp/1107694167) 188 | 189 | * Causal Inference for Statistics, Social, and Biomedical Sciences. 190 | An Introduction, Imbens & Rubin, (2015). 191 | [amzn](https://www.amzn.com/dp/0521885884). 192 | 193 | * Why ask Why? Forward Causal Inference and Reverse Causal Questions. 194 | Andrew Gelman & Guido Imbens 195 | [doi](http://doi.org/10.3386/w19614) (2013) 196 | 197 | * Inference and Intervention: Causal Models for Business Analysis. 198 | Michael D. Ryall and Aaron L. Bramson (2013). 199 | [amzn](https://www.amzn.com/dp/0415657598) 200 | 201 | * Angrist & Pischke. 202 | Mostly Harmless Econometrics (2009). 203 | [amzn](https://www.amzn.com/dp/0691120358/). 204 | [princeton](https://press.princeton.edu/titles/8769.html) 205 | 206 | * Probabilistic Graphical Models: Principles and Techniques. 207 | Daphne Koller and Nir Friedman. (2009). 208 | MIT Press. 209 | [amzn](https://www.amzn.com/dp/0262013193). 210 | [mit](https://mitpress.mit.edu/books/probabilistic-graphical-models) 211 | 212 | * Judea Perl. 213 | Causality: Models, Reasoning and Inference (2009) 2nd Edition. 214 | [amzn](https://www.amzn.com/dp/052189560X). 215 | 216 | * Fundamentals of statistical causality (2007). 217 | C. A. P. Dawid. 218 | [pre-print research gate](https://www.researchgate.net/publication/242495222_Fundamentals_of_statistical_causality). 219 | 220 | * Microeconometrics: Methods and Applications (2005). 221 | Cameron, A. C., and P. K. Trivedi. 222 | Cambridge: Cambridge University Press. [amz](https://www.amzn.com/dp/0521848059). 223 | Chapter-4: Instrumental Variables section [Author's-copy](http://cameron.econ.ucdavis.edu/e240a/ch04iv.pdf). 224 | 225 | * Rosenbaum. 226 | Observational Studies (Springer Series in Statistics) 2nd Edition (2002). 227 | [amzn](https://www.amzn.com/dp/0387989676) 228 | 229 | * Causation, Prediction, and Search. 230 | Peter Spirtes, Clark Glymour and Richard Scheines. 231 | 2nd Edition (2001). 232 | [mit press](https://mitpress.mit.edu/books/causation-prediction-and-search-second-edition) | [pdf-research-gate](https://www.researchgate.net/publication/242448131_Causation_Prediction_and_Search) 233 | 234 | * Structural Equations with Latent Variables. 235 | Kenneth A. Bollen (1989). 236 | [wiley](https://onlinelibrary.wiley.com/doi/book/10.1002/9781118619179). 237 | [amzn](https://www.amzn.com/dp/0471011711). 238 | 239 | 240 | ## Papers general 241 | 242 | * Causal inference for time series 243 | Jakob Runge et. al. 244 | Nature Reviews Earth & Environment volume 4, pages 487–505 (2023) 245 | [doi](https://doi.org/10.1038/s43017-023-00431-y) 246 | 247 | * A Measure-Theoretic Axiomatisation of Causality. 248 | Junhyung Park, Simon Buchholz, Bernhard Schölkopf, Krikamol Muandet. 249 | [arXiv](https://arxiv.org/abs/2305.17139) (2023). 250 | 251 | * Double Machine Learning and Automated Confounder Selection -- A Cautionary Tale. 252 | Paul Hünermund et. al. (2023) 253 | [arXiv](https://arxiv.org/abs/2108.11294) | [talk](https://faculti.net/double-machine-learning-and-automated-confounder-selection/) | 254 | [Journal of Causal Inference](https://doi.org/10.1515/jci-2022-0078) 255 | 256 | * Causal Inference and Data-Fusion in Econometrics. 257 | Paul Hünermund (Maastricht University), Elias Bareinboim (Columbia University). 258 | [arXiv:1912.09104](https://arxiv.org/abs/1912.09104). | [The Econometrics Journal, 2023](https://doi.org/10.1093/ectj/utad008) 259 | 260 | * Phenomenological Causality. 261 | Dominik Janzing, Sergio Hernan Garrido Mejia (2022). 262 | [arXiv](https://arxiv.org/abs/2211.09024) 263 | 264 | * Personalized Decision Making -- A Conceptual Introduction 265 | Scott Mueller, Judea Pearl. 266 | [arXiv:2208.09558](https://arxiv.org/abs/2208.09558). (2022). 267 | 268 | * Backtracking Counterfactuals. 269 | Julius von Kügelgen, Abdirisak Mohamed, Sander Beckers. 270 | [arXiv](https://arxiv.org/abs/2211.00472) (2022) 271 | 272 | * Causal Entropy Optimization. 273 | Nicola Branchini, Virginia Aglietti, Neil Dhir, Theodoros Damoulas. 274 | [arXiv](https://arxiv.org/abs/2208.10981). (2022). 275 | 276 | * From Statistical to Causal Learning. 277 | Bernhard Schölkopf, Julius von Kügelgen 278 | [arXiv](https://arxiv.org/abs/2204.00607) (2022) 279 | 280 | * Sensitivity Analysis of Individual Treatment Effects: A Robust Conformal Inference Approach. 281 | Ying Jin, Zhimei Ren, Emmanuel J. Candès. (2021). 282 | [arXiv](https://arxiv.org/abs/2111.12161). | [Rpackage](https://zhimeir.github.io/cfsensitivity/index.html) 283 | 284 | * A Survey on Causal Inference. 285 | Yao et. al. 286 | [doi](https://dl.acm.org/doi/10.1145/3444944). (2021). 287 | 288 | * Potential Outcome and Directed Acyclic Graph Approaches to Causality: 289 | Relevance for Empirical Practice in Economics. 290 | Guido W. Imbens. 291 | Journal of Economic Literature (December 2020). 292 | [journal](https://www.aeaweb.org/articles?id=10.1257/jel.20191597) | [arXiv](https://arxiv.org/abs/1907.07271) | [Pearl's response](http://causality.cs.ucla.edu/blog/index.php/category/imbens/) 293 | 294 | * Synthetic Difference in Differences, 2019. 295 | Dmitry Arkhangelsky, Susan Athey, David A. Hirshberg, Guido W. Imbens, and Stefan Wager. 296 | [arxiv:1812.09970](https://arxiv.org/abs/1812.09970) 297 | 298 | * Why Propensity Scores Should Not Be Used for Matching 299 | King et. al. 300 | Political Analysis, 27, 4, Pp. 435-454. (2019) 301 | [url](https://tinyurl.com/y5b5yjxo) 302 | 303 | * Detecting and quantifying causal associations in large nonlinear time series datasets. 304 | Jakob Runge, Peer Nowack, Marlene Kretschmer, Seth Flaxman and Dino Sejdinovic. 305 | Science Advances 27 Nov 2019 Vol. 5, no. 11, eaau4996. 306 | [doi](http://dx.doi.org10.1126/sciadv.aau4996) 307 | 308 | * The Seven Tools of Causal Inference, with Reflections on Machine Learning. 309 | Judea Pearl. 310 | Communications of the ACM, March 2019, Vol. 62 No. 3, Pages 54-60. 311 | [url](https://cacm.acm.org/magazines/2019/3/234929-the-seven-tools-of-causal-inference-with-reflections-on-machine-learning/fulltext) 312 | [pdf-reprint](https://ftp.cs.ucla.edu/pub/stat_ser/r481-reprint.pdf). 313 | [interview](https://www.youtube.com/watch?v=CsMV5o3hotY). 314 | 315 | * A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms. 316 | Yoshua Bengio, Tristan Deleu, Nasim Rahaman, Rosemary Ke, Sébastien Lachapelle, 317 | Olexa Bilaniuk, Anirudh Goyal, Christopher Pal. 318 | [arXiv:1901.10912](https://arxiv.org/abs/1901.10912) 319 | 320 | * Theoretical Impediments to Machine Learning With Seven Sparks from the Causal Revolution. 321 | Judea Pearl. 322 | [arXiv:1801.04016](https://arxiv.org/abs/1801.04016) 323 | 324 | * Automated versus do-it-yourself methods for causal inference: 325 | Lessons learned from a data analysis competition. 326 | Vincent Dorie†, Jennifer Hill, Uri Shalit, Marc Scott, and Dan Cervone. 327 | [arXiv:1707.02641](https://arxiv.org/pdf/1707.02641.pdf) 328 | 329 | * Double/Debiased Machine Learning for Treatment and Causal Parameters. 330 | Victor Chernozhukov et. al. (2017). 331 | [arXiv](https://arxiv.org/abs/1608.00060). | [EC-journal](https://academic.oup.com/ectj/article/21/1/C1/5056401?login=true) 332 | 333 | * Quantum-Like Bayesian Networks for Modeling Decision Making. 334 | Catarina Moreira and Andreas Wichert. 335 | Front. Psychol., 26 January 2016. 336 | [doi](https://doi.org/10.3389/fpsyg.2016.00011) 337 | 338 | * Causal inference and the data-fusion problem 339 | Elias Bareinboim and Judea Pearl 340 | PNAS, 113 (27) 7345-7352 (2016) 341 | [doi](https://doi.org/10.1073/pnas.1510507113) 342 | 343 | * The Sure-Thing Principle. 344 | Judea Pearl (2016). 345 | [pdf](https://ftp.cs.ucla.edu/pub/stat_ser/r466.pdf) 346 | 347 | * Brodersen KH, Gallusser F, Koehler J, Remy N, Scott SL. 348 | Inferring causal impact using Bayesian structural time-series models. 349 | Annals of Applied Statistics, (2015), Vol. 9, No. 1, 247-274. 350 | [url](http://research.google.com/pubs/pub41854.html) 351 | 352 | * Linear Models: A Useful “Microscope” for Causal Analysis. 353 | Journal of Causal Inference 2013; 1(1): 155–170. 354 | Judea Pearl. 355 | [doi](https://doi.org/10.1515/jci-2013-0003) [pdf](http://ftp.cs.ucla.edu/pub/stat_ser/r409-corrected-reprint.pdf) 356 | 357 | * Introduction to Judea Pearl's Do-Calculus. 358 | Robert R. Tucci. (2013) 359 | [arXiv:1305.5506](https://arxiv.org/abs/1305.5506) 360 | 361 | * The Do-Calculus Revisited. 362 | Judea Pearl. 363 | Keynote AI Uncertainy Conference (2012). 364 | [pdf](https://ftp.cs.ucla.edu/pub/stat_ser/r402.pdf) 365 | 366 | * Detecting causality in complex ecosystems. 367 | George Sugihara, Robert May, Hao Ye, Chih-hao Hsieh, Ethan Deyle, Michael Fogarty, Stephan Munch. 368 | Science Oct 26;338(6106):496-500 (2012). 369 | [doi](https://doi.org/10.1126/science.1227079). 370 | `convergent cross-mapping` 371 | 372 | * Introduction to Causal Inference. 373 | Peter Spirtes. 374 | (2010) 375 | [jmlr](http://www.jmlr.org/papers/v11/spirtes10a.html). 376 | 377 | * Transfer entropy—a model-free measure of effective connectivity for the neurosciences. 378 | Vicente et. al. 379 | J Comput Neurosci (2011) 30:45–67. 380 | [pdf](https://link.springer.com/content/pdf/10.1007/s10827-010-0262-3.pdf). 381 | 382 | * Causal inference in statistics: An overview. 383 | Judea Pearl. (2009). 384 | [doi](http://dx.doi.org/10.1214/09-SS057). 385 | 386 | * The Neyman— Rubin Model of Causal Inference and Estimation Via Matching Methods. 387 | Jasjeet Sekhon. 388 | The Oxford Handbook of Political Methodology. 389 | [doi](10.1093/oxfordhb/9780199286546.003.0011) | [pdf-preprint](http://sekhon.berkeley.edu/papers/SekhonOxfordHandbook.pdf) 390 | 391 | * Econometric Causality. 392 | James J. Heckman. 393 | International Statistical Review (2008), 76, 1, 1–27. 394 | [doi](http://dx.doi.org/10.1111/j.1751-5823.2007.00024.x). 395 | 396 | * Causal Inference Using Potential Outcomes: Design, Modeling, Decisions. 397 | Donald B. Rubin. 398 | Journal of the American Statistical Association. 399 | Vol. 100, No. 469 (Mar., 2005), pp. 322-331. 400 | [jstor](https://www.jstor.org/stable/27590541). 401 | [pdf-tandfonline](https://www.tandfonline.com/doi/pdf/10.1198/016214504000001880?casa_token=uQtd_czuZDEAAAAA:uh_ziMbrAJflEo1a5JbkhZauwazlpmdJKkbo0zRHRBlzBWx7CnS4GZAttMPyOVqow70mUycX0lmwgg) 402 | 403 | * The Economic Costs of Conflict: A Case Study of the Basque Country 404 | Alberto Abadie, Javier Gardeazabal 405 | American Economic Review, Vol. 93, No.1, pp 113-132 (2003) | 406 | [doi](https://dx.doi.org/10.1257/000282803321455188) | [baylor-pdf](https://business.baylor.edu/scott_cunningham/teaching/abadie-and-gardeazabal-2003.pdf) 407 | 408 | * Time Series Analysis, Cointegration, and applications. 409 | Nobel Lecture of Clive Granger. 410 | (2003) 411 | [pdf](https://www.nobelprize.org/uploads/2018/06/granger-lecture.pdf) 412 | 413 | * Causal Complexity and the Study of Politics. 414 | Bear Braumoeller. 415 | Political Analysis 198(11):209-233 (2003). 416 | [doi](http://dx/doi.org/10.1093/pan/mpg012). 417 | [researchgate-url](https://www.researchgate.net/publication/228774938_Causal_Complexity_and_the_Study_of_Politics). 418 | 419 | * Identification of Causal Effects Using Instrumental Variables 420 | Joshua D. Angrist, Guido W. Imbens and Donald B. Rubin 421 | Journal of the American Statistical Association (JASA) 422 | Vol. 91, No. 434 (Jun., 1996) [jstor](https://www.jstor.org/stable/2291629) | [pdf-McGill](https://www.math.mcgill.ca/dstephens/AngristIV1996-JASA-Combined.pdf) 423 | `potential outcomes` 424 | 425 | * Causal diagrams for empirical research. 426 | Judea Pearl (1995). 427 | [jstor](https://www.jstor.org/stable/2337329) | [pdf-UCLA](http://bayes.cs.ucla.edu/R218-B.pdf). 428 | `Reasoning on Graphs: d-seperation, back/front-door` 429 | 430 | * Identification and Estimation of Local Average Treatment Effects. 431 | Joshua D. Angrist & Guido W. Imbens. 432 | [nber](https://www.nber.org/papers/t0118) (1995). 433 | Econometrica, 6, 2,467-476. (1994) [jstor](https://www.jstor.org/stable/2951620) | [pdf-Baylor](https://business.baylor.edu/scott_cunningham/teaching/imbens--angrist---late-1994.pdf) 434 | 435 | * Paul W. Holland. 436 | Statistics and Causal Inference. 437 | Journal of the American Statistical Association. 438 | Dec., 1986, Vol. 81, No. 396 (Dec., 1986), pp. 945-960 [jstor](https://www.jstor.org/stable/2289064). | [pdf-from-David-Bleie](https://www.cs.columbia.edu/~blei/fogm/2020F/readings/Holland1986.pdf) 439 | 440 | * Rubin, D. B. (1974) 441 | Estimating causal effects of treatments in randomized and nonrandomized studies. 442 | Journal of Educational Psychology, 66(5), 688-701. 443 | [doi](http://dx.doi.org/10.1037/h0037350) 444 | 445 | * Haavelmo, T. (1943). 446 | The statistical implications of a system of simultaneous equations. 447 | Econometrica, 11, 1–12. 448 | [jstor](http://links.jstor.org/sici?sici=0012-9682%28194301%2911%3A1%3C1%3ATSIOAS%3E2.0.CO%3B2-N) 449 | 450 | Judea Pearls comment on Haavelmo. 451 | Econometric Theory , Volume 31 , Issue 1: Haavelmo Memorial Issue: Part One , February 2015 , pp. 152 - 179 452 | [doi](https://doi.org/10.1017/S0266466614000231). 453 | 454 | * Wright, S. (1921) 455 | Correlation and causation. 456 | J. Agricultural Research. 20: 557–585. 457 | 458 | * Wright, S. (1934). 459 | The method of path coefficients. 460 | Annals of Mathematical Statistics. 5 (3): 161–215. 461 | [doi](http://www.doi.org/10.1214/aoms/1177732676) 462 | 463 | ## Causal Discovery 464 | 465 | * Causation versus Prediction: Comparing Causal Discovery and 466 | Inference with Artificial Neural Networks in Travel Mode Choice Modeling 467 | Rishabh Singh Chauhan, Uttara Sutradhar, Anton Rozhkov, Sybil Derrible 468 | [arXiv](https://arxiv.org/abs/2307.15262) (2023) 469 | `Comparison of PC and Double ML on transportation preference` 470 | 471 | * An Introduction to Causal Discovery 472 | Martin Huber, A bonus chapter (2023) 473 | [pdf-academic-page](https://www.unifr.ch/appecon/en/assets/public/uploads/introcausaldiscovery2.pdf) 474 | 475 | * Causal Discovery and Prediction: Methods and Algorithms 476 | Gilles Blondel (Thesis) (2023) 477 | [arXiv](https://arxiv.org/abs/2309.09416) 478 | 479 | * On the Interventional Kullback-Leibler Divergence. 480 | Jonas Wildberger, Siyuan Guo, Arnab Bhattacharyya, Bernhard Schölkopf. 481 | [arXiv](https://arxiv.org/abs/2302.05380) (2023). 482 | 483 | * Causal Structure Learning: a Combinatorial Perspective. 484 | Chandler Squires, Caroline Uhler. (2022). 485 | [arXiv](https://arxiv.org/abs/2206.01152). 486 | 487 | * Deep End-to-end Causal Inference. 488 | Tomas Geffner, Javier Antoran, Adam Foster, Wenbo Gong, 489 | Chao Ma, Emre Kiciman, Amit Sharma, Angus Lamb, Martin Kukla, 490 | Nick Pawlowski, Miltiadis Allamanis, Cheng Zhang. 491 | [arXiv](https://arxiv.org/abs/2202.02195) (2022). 492 | [github-deci](https://github.com/microsoft/project-azua/tree/main/azua/models/deci) 493 | 494 | * Learning to Induce Causal Structure. 495 | Nan Rosemary Ke, Silvia Chiappa, Jane Wang, Jorg Bornschein, Theophane Weber, 496 | Anirudh Goyal, Matthew Botvinic, Michael Mozer, Danilo Jimenez Rezende. 497 | [arXiv](https://arxiv.org/abs/2204.04875) (2022). 498 | 499 | * Discovering Causal Structure with Reproducing-Kernel Hilbert Space ε-Machines. 500 | Nicolas Brodu, James P. Crutchfield. 501 | [arXiv](https://arxiv.org/abs/2011.14821) (2021). 502 | 503 | * Nobel Memorial Economics Prize 2021 on causal discovery, scientific summary. 504 | Answering Causal Questions Using Observational Data. (2021) 505 | (David Card, Joshua Angrist, and Guido Imbens) 506 | [pdf](https://www.nobelprize.org/uploads/2021/10/advanced-economicsciencesprize2021.pdf). 507 | 508 | * Causal network reconstruction from time series: From theoretical assumptions to practical estimation. 509 | J. Runge. 510 | Chaos 28, 075310 (2018). 511 | [DOI](https://doi.org/10.1063/1.5025050). 512 | 513 | * Learning Representations for Counterfactual Inference. 514 | Fredrik D. Johansson, Uri Shalit, David Sontag. 515 | ICML 2016, [arXiv](https://arxiv.org/abs/1605.03661). 516 | 517 | * Causal Discovery via Reproducing Kernel Hilbert Space Embeddings 518 | Zhitang Chen, Kun Zhang, Laiwan Chan, Bernhard Schölkopf 519 | Neural Computation 26, 1484–1517 (2014). 520 | [doi](http://dx.doi.org/10.1162/NECO_a_00599) 521 | 522 | * Cosma Shalizi's notebook [causal discovery algorithms](http://bactra.org/notebooks/causal-discovery-algorithms.html). 523 | 524 | ## Simpsons Paradox 525 | 526 | * Learning to Discover Various Simpson's Paradoxes 527 | Jingwei Wang, Jianshan He, Weidi Xu, Ruopeng Li, Wei Chu 528 | KDD '23: Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining 529 | August 2023, Pages 5092–5103 [doi](https://doi.org/10.1145/3580305.3599859) | [git-repository](https://github.com/ant-research/Learning-to-Discover-Various-Simpson-Paradoxes) 530 | 531 | * A toolbox to demystify probabilistic and statistical paradoxes. 532 | Kelter R, Schnurr A and Spies S (2023) 533 | Front. Educ. 8:1212419. [doi](https://doi.org/10.3389/feduc.2023.1212419) | [author-pdf](https://dspace.ub.uni-siegen.de/bitstream/ubsi/2578/3/A_toolbox_to_demystify_probabilistic_and_statistical_paradoxes.pdf) 534 | 535 | * Pearl, J. (2014). Comment: understanding Simpson’s paradox. Am. Stat. 68, 8–13. doi: 10.1080/00031305.2014.876829 536 | 537 | * Simpson's Paradox: An Anatomy 538 | Judea Pearl (1999). 539 | [UCLA Technical-Report](http://bayes.cs.ucla.edu/R264.pdf) 540 | 541 | * On Simpson's Paradox and the Sure-Thing Principle. 542 | Colin R. Blyth. 543 | Journal of the American Statistical Association. 544 | Vol. 67, No. 338 (Jun., 1972), pp. 364-366. [doi](https://doi.org/10.2307/2284382) 545 | 546 | * The interpretation of interaction in contingency tables. 547 | Simpson, E. H. (1951). 548 | J. Royal Stat. Soc. 13, 238–241. [doi](https://dx.doi.org/10.1111/J.2517-6161.1951.TB00088.X) 549 | 550 | ## Machine Learning 551 | Including games, reinforcement or deep learning. 552 | 553 | * Comprehensive Causal Machine Learning 554 | Michael Lechner, Jana Mareckova 555 | [arxiv](https://arxiv.org/abs/2405.10198)| (2024) 556 | 557 | * Language Models as Causal Effect Generators 558 | Lucius E.J. Bynum, Kyunghyun Cho 559 | [arXiv](https://arxiv.org/abs/2411.08019) (2024) 560 | 561 | * Invariant Causal Knowledge Distillation in Neural Networks 562 | Nikolaos Giakoumoglou, Tania Stathaki 563 | [arXiv](https://arxiv.org/abs/2407.11802v2) (2024) 564 | 565 | * Peer-induced Fairness: A Causal Approach for Algorithmic Fairness Auditing 566 | Shiqi Fang, Zexun Chen, Jake Ansell 567 | [arXiv](https://arxiv.org/abs/2408.02558) (2024) 568 | 569 | * Causally Abstracted Multi-armed Bandits 570 | Fabio Massimo Zennaro et. al. 571 | [arXiv](https://arxiv.org/abs/2404.17493) (2024) 572 | 573 | * Causal machine learning for predicting treatment outcomes 574 | Stefan Feuerriegel et. al. 575 | [url](https://www.nature.com/articles/s41591-024-02902-1) (2023) 576 | 577 | * Causal machine learning for single-cell genomics 578 | Alejandro Tejada-Lapuerta, Paul Bertin, Stefan Bauer, Hananeh Aliee, Yoshua Bengio, Fabian J. Theis 579 | [arXiv](https://arxiv.org/abs/2310.14935) (2023) 580 | 581 | * Estimating categorical counterfactuals via deep twin networks 582 | Athanasios Vlontzos, Bernhard Kainz & Ciarán M. Gilligan-Lee 583 | Nature Machine Intelligence volume 5, pages 159–168 (2023) 584 | [nature](https://www.nature.com/articles/s42256-023-00611-x) | [arXiv](https://arxiv.org/abs/2109.01904) 585 | 586 | * Causal Reasoning and Large Language Models: Opening a New Frontier for Causality 587 | Emre Kıcıman, Robert Ness, Amit Sharma, Chenhao Tan 588 | [arXiv](https://arxiv.org/abs/2305.00050) (2023) 589 | 590 | * Causal Reinforcement Learning: A Survey 591 | Zhihong Deng, Jing Jiang, Guodong Long, Chengqi Zhang 592 | [arXiv](https://arxiv.org/abs/2307.01452) (2023) 593 | 594 | * Causal Component Analysis. 595 | Wendong Liang et. al. 596 | [arXiv](https://arxiv.org/abs/2305.17225) (2023). 597 | 598 | * Double Machine Learning and Automated Confounder Selection -- A Cautionary Tale. 599 | Paul Hünermund et. al. (2023) 600 | [arXiv](https://arxiv.org/abs/2108.11294) | [talk](https://faculti.net/double-machine-learning-and-automated-confounder-selection/) | 601 | [Journal of Causal Inference](https://doi.org/10.1515/jci-2022-0078) 602 | 603 | * Causal Deep Learning. 604 | Jeroen Berrevoets, Krzysztof Kacprzyk, Zhaozhi Qian, Mihaela van der Schaar. (2023) 605 | [arXiv](https://arxiv.org/abs/2303.02186). 606 | 607 | * A Survey on Causal Reinforcement Learning 608 | Yan Zeng et. al. [arXiv](https://arxiv.org/abs/2302.05209) (2023) 609 | 610 | * Reasoning about Causality in Games. 611 | Lewis Hammond, James Fox, Tom Everitt, Ryan Carey, Alessandro Abate, Michael Wooldridge. (2023). 612 | [arXiv](https://arxiv.org/abs/2301.02324). 613 | 614 | * Evaluating Uses of Deep Learning Methods for Causal Inference. 615 | Albert Whata; Charles Chimedza. 616 | [ieee](https://ieeexplore.ieee.org/abstract/document/9667520). (2022) 617 | 618 | * Causal Machine Learning: A Survey and Open Problems 619 | Jean Kaddour, Aengus Lynch, Qi Liu, Matt J. Kusner, Ricardo Silva 620 | [arXiv](https://arxiv.org/abs/2206.15475) (2022) 621 | 622 | * Causal Reinforcement Learning using Observational and Interventional Data. 623 | Maxime Gasse, Damien Grasset, Guillaume Gaudron, Pierre-Yves Oudeyer. 624 | [arXiv:2106.14421](https://arxiv.org/abs/2106.14421) (2021) 625 | 626 | * Causal Reinforcement Learning, ICML 2020 [url](https://crl.causalai.net) 627 | 628 | * Sample-Efficient Reinforcement Learning via Counterfactual-Based Data Augmentation. 629 | Chaochao Lu, Biwei Huang, Ke Wang, José Miguel Hernández-Lobato, Kun Zhang, Bernhard Schölkopf. (2020). 630 | [arXiv:2012.09092](https://arxiv.org/abs/2012.09092) 631 | 632 | * Machine Learning Methods That Economists Should Know About 633 | Susan Athey and Guido W. Imbens 634 | [Annual Reviews](https://www.annualreviews.org/content/journals/10.1146/annurev-economics-080217-053433;jsessionid=ln_r7UHAEX_IEsVFDhG1ZgtVXrsPhWv5ReUKrihr.ip-10-241-10-70) (2019) 635 | 636 | * Causality for Machine Learning. 637 | Bernhard Schölkopf. 638 | [arXiv:1911.10500](https://arxiv.org/abs/1911.10500) 639 | 640 | * Causal Inference and Uplift Modelling: A Review of the Literature. 641 | Pierre Gutierrez, Jean-Yves Gérardy ; Proceedings of The 3rd International 642 | Conference on Predictive Applications and APIs, PMLR 67:1-13, 2017. 643 | [url](http://proceedings.mlr.press/v67/gutierrez17a.html) 644 | 645 | ## Fairness 646 | 647 | * Survey on Causal-based Machine Learning Fairness Notions. 648 | Karima Makhlouf, Sami Zhioua, Catuscia Palamidessi. 649 | [arXiv](https://zhimeir.github.io/cfsensitivity/index.html) (2020). 650 | 651 | * Causal Modeling of Twitter Activity during COVID-19 652 | Gencoglu, Oguzhan, and Mathias Gruber. (2020). 653 | Computation 8, no. 4: 85. [doi](https://doi.org/10.3390/computation8040085). 654 | 655 | * Achieving Causal Fairness in Machine Learning. 656 | Wu, Yongkai. University of Arkansas, ProQuest Dissertations Publishing, 2020. 27960037. 657 | [url](https://search.proquest.com/openview/5b94283bd4da8edc1b14bff3db4c9e77/1?pq-origsite=gscholar&cbl=18750&diss=y) 658 | 659 | * Causal Bayesian Networks (2019). 660 | [blog-deepmind](https://deepmind.com/blog/article/Causal_Bayesian_Networks). 661 | 662 | * A Causal Bayesian Networks Viewpoint on Fairness. 663 | Silvia Chiappa, William S. Isaac (2019). 664 | [arXiv:1907.06430](https://arxiv.org/abs/1907.06430) 665 | 666 | * Zhang, J., & Bareinboim, E. (2018). 667 | Fairness in Decision-Making — The Causal Explanation Formula. 668 | Proceedings of the AAAI Conference on Artificial Intelligence, 32(1). 669 | [doi](https://doi.org/10.1609/aaai.v32i1.11564). 670 | 671 | * Counterfactual Fairness. 672 | Kusner et. al. 673 | NeurIPs 2017 [pdf](https://proceedings.neurips.cc/paper/2017/file/a486cd07e4ac3d270571622f4f316ec5-Paper.pdf) 674 | 675 | * Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPR 676 | Sandra Wachter, Brent Mittelstadt, Chris Russell 677 | [arXiv](https://arxiv.org/abs/1711.00399) (2017). 678 | 679 | ## Physics 680 | 681 | * H-Theorem do-conjecture 682 | M. Suzen 683 | [arXiv:2310.01458](https://arxiv.org/abs/2310.01458). | [code](https://github.com/msuzen/research/tree/main/h-do-conjecture) 684 | 685 | * Causality, determinism, and physics. 686 | Julio Geo-Banacloche. 687 | American Journal of Physics 90, 809 (2022); [doi](https://doi.org/10.1119/5.0087017). 688 | 689 | * Arrow of Causality and Quantum Gravity. 690 | John F. Donoghue and Gabriel Menezes. 691 | Phys. Rev. Lett. 123, 171601 692 | [url](https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.123.171601) 693 | 694 | * The causal set approach to quantum gravity. 695 | Sumati Surya 696 | Living Reviews in Relativity volume 22, Article number: 5 (2019). 697 | [url](https://link.springer.com/article/10.1007/s41114-019-0023-1). 698 | 699 | * Algorithmic independence of initial condition and dynamical law in thermodynamics and causal inference. 700 | Dominik Janzing et al 2016 New J. Phys. 18 093052. 701 | [url-open-access](https://iopscience.iop.org/article/10.1088/1367-2630/18/9/093052/meta). 702 | 703 | * Quantum causality. 704 | Brukner, Časlav. 705 | Nature Physics, Volume 10, Issue 4, pp. 259-263 (2014). 706 | [doi](https://ui.adsabs.harvard.edu/link_gateway/2014NatPh..10..259B/doi:10.1038/nphys2930) | [yale-pdf](https://inferenceproject.yale.edu/sites/default/files/brukner2014_q_causality.pdf) 707 | 708 | * Causal Entropic Forces. 709 | A. D. Wissner-Gross and C. E. Freer. 710 | Phys. Rev. Lett. 110, 168702. (2013). 711 | [url](https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.110.168702). 712 | 713 | * Resource Letter CD-1: Causality and determinism in physics. 714 | R. Jones. 715 | Am. J. Phys. 64, 208–215 (1996). [doi](https://doi.org/10.1119/1.18208). 716 | 717 | * Space-time as a causal set. 718 | Luca Bombelli, Joohan Lee, David Meyer, and Rafael D. Sorkin. 719 | Phys. Rev. Lett. 59, 521 (1987). 720 | [url](https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.59.521) 721 | 722 | * The class of continuous timelike curves determines the topology of spacetime. 723 | David B. Malament. 724 | Journal of Mathematical Physics 18, 1399 (1977). 725 | [doi](https://doi.org/10.1063/1.523436). 726 | 727 | 728 | ## Thesis 729 | 730 | PhD thesis on causality. 731 | 732 | * Causality in Point Processes 733 | McGovern, Ian 734 | PhD Thesis, Los Angeles, [url](https://escholarship.org/content/qt5x56b2fk/qt5x56b2fk.pdf) (2024) 735 | 736 | * Identifiable Causal Representation Learning 737 | Unsupervised, Multi-View, and Multi-Environment 738 | Julius von Kügelgen 739 | PhD Thesis, Cambridge [doi](https://doi.org/10.17863/CAM.106852) (2023) 740 | 741 | * Explainable Reinforcement Learning Through a Causal Lens. 742 | Prashan Madumal. 743 | PhD thesis, Melbourne [pdf](https://rest.neptune-prod.its.unimelb.edu.au/server/api/core/bitstreams/0730665b-be59-5e97-9eb8-33223bf6464c/content) | [aaai](https://ojs.aaai.org//index.php/AAAI/article/view/5631). (2021). 744 | 745 | ## Review 746 | 747 | * Causal Inference in the Social Sciences 748 | Guido W. Imbens 749 | [url](https://www.annualreviews.org/content/journals/10.1146/annurev-statistics-033121-114601) (2024) 750 | 751 | ## Software. 752 | 753 | ### Causal Discovery 754 | 755 | * Causal-learn: Causal Discovery in Python 756 | Spirtes & CMU & Team 757 | [arXiv](https://arxiv.org/abs/2307.16405) | [repo](https://github.com/py-why/causal-learn) 758 | 759 | * Tigramite – Causal inference and causal discovery for time series datasets. 760 | [github](https://github.com/jakobrunge/tigramite) 761 | 762 | * The Causal Discovery Toolbox (CDT) 763 | A package for causal inference in graphs and in the pairwise settings. 764 | [github](https://github.com/FenTechSolutions/CausalDiscoveryToolbox). | [arXiv](https://arxiv.org/abs/1903.02278) 765 | 766 | * LinGAM Discovery of non-gaussian linear causal models [github](https://github.com/cdt15/lingam) | [software paper](https://jmlr.org/papers/v24/21-0321.html) | [original paper](https://www.jmlr.org/papers/v7/shimizu06a.html) 767 | 768 | * PyPhi: A toolbox for integrated information theory. 769 | [pypi](https://pypi.org/project/pyphi/). 770 | [arXiv1712.09644](https://arxiv.org/abs/1712.09644) 771 | 772 | * Causal-tune [github](https://github.com/py-why/causaltune). 773 | 774 | * Huawei's gCastle is a causal structure learning toolchain [github-repo](https://github.com/huawei-noah/trustworthyAI/tree/master/gcastle) 775 | 776 | ### Microsoft Research : PyWhy 777 | 778 | * PyWhy organisation [github](https://github.com/py-why/). 779 | [DoWhy evolves to independent PyWhy model to help causal inference grow](https://www.microsoft.com/en-us/research/blog/dowhy-evolves-to-independent-pywhy-model-to-help-causal-inference-grow/?OCID=msr_blog_PyWhy_TW) | [aws-blog](https://www.amazon.science/blog/aws-contributes-novel-causal-machine-learning-algorithms-to-dowhy) 780 | 781 | * DoWhy is a Python library for causal inference. 782 | [github](https://github.com/microsoft/dowhy) | [Paper-arXiv](https://arxiv.org/abs/2011.04216). 783 | 784 | * EconML. 785 | EconML is a Python package for estimating heterogeneous treatment 786 | effects from observational data via machine learning. 787 | [github](https://github.com/microsoft/EconML) 788 | 789 | * Azua/DECI. 790 | [github](https://github.com/microsoft/project-azua) 791 | [github-DECI-module](https://github.com/microsoft/project-azua/tree/main/azua/models/deci) `Deep End-to-end Causal Inference` 792 | 793 | * ShowWhy. 794 | UI on top of DoWhy, EconML [github](https://github.com/microsoft/showwhy) 795 | 796 | * Causica : DECI: End to End Causal Inference 797 | [github](https://github.com/microsoft/causica) 798 | 799 | * Causal-learn: Causal Discovery in Python 800 | Spirtes & CMU & Team 801 | [arXiv](https://arxiv.org/abs/2307.16405) | [repo](https://github.com/py-why/causal-learn) 802 | 803 | * llm experimental [pywhy-llm](https://github.com/py-why/pywhy-llm) 804 | 805 | ### Causal Impact and Observational 806 | 807 | * DoubleML in Python/R package [github](https://github.com/DoubleML/doubleml-for-py). | [JSS-article](https://www.jstatsoft.org/article/view/v108i03) 808 | 809 | * CausalML: A Python Package for Uplift Modeling and Causal Inference with ML. 810 | [github](https://github.com/uber/causalml) | [software X paper](https://doi.org/10.1016/j.softx.2022.101294) 811 | 812 | * An R package for causal inference using Bayesian structural time-series models 813 | [CausalImpact](https://google.github.io/CausalImpact/CausalImpact.html). 814 | [CRAN](https://cran.r-project.org/package=CausalImpact). 815 | [github-python-port](https://github.com/jamalsenouci/causalimpact). 816 | [paper](https://research.google/pubs/pub41854/). 817 | [causalimpact-tf](https://github.com/WillianFuks/tfcausalimpact) Re-write with tensorflow probability 818 | 819 | * upliftml : Uplift modelling, Booking.com 820 | [github](https://github.com/bookingcom/upliftml) 821 | 822 | * [Identifying Causal Effects with the R Package causaleffect](https://cran.r-project.org/web/packages/causaleffect/vignettes/causaleffect.pdf) 823 | 824 | * IBM's causalib Python package [github](https://github.com/IBM/causallib) 825 | 826 | * (synthdid: Synthetic Difference in Differences Estimation)[https://synth-inference.github.io/synthdid/index.html]. 827 | R-package 828 | 829 | ### Causal Graphs and Bayesian Networks 830 | 831 | * DAGitty — draw and analyze causal diagrams [url](https://www.dagitty.net) 832 | 833 | * CausalQueries: Make, Update, and Query Binary Causal Models 834 | [CRAN](https://cran.rstudio.com/web/packages/CausalQueries/index.html) | [book: Causal Models: Guide to CausalQueries](https://macartan.github.io/causalmodels/) 835 | 836 | * pgmpy is a pure python implementation for Bayesian Networks [www](https://pgmpy.org) | [Paper-ArXiv](https://arxiv.org/abs/2304.08639) | 837 | 838 | * causaleffect: Deriving Expressions of Joint Interventional Distributions and Transport Formulas in Causal Models. 839 | [CRAN](https://cran.r-project.org/web/packages/causaleffect/index.html). 840 | * GRAPHL [repo](https://github.com/max-little/GRAPL) | [joss](https://joss.theoj.org/papers/10.21105/joss.04534.pdf) 841 | 842 | * dagR: Directed Acyclic Graph with R. 843 | [CRAN](https://cran.r-project.org/web/packages/dagR/index.html)[doi](http://dx.doi.org/10.1097/EDE.0b013e3181e09112) 844 | 845 | * PyCID: Causal Influence Diagrams library [github](https://github.com/causalincentives/pycid) 846 | Relevant to PyCID : A python library for 2 player games [nashpy](https://github.com/drvinceknight/nashpy) 847 | 848 | * [CausalPy](https://github.com/pymc-labs/CausalPy) : Bayesian -regression discontinuity | 849 | [pymc-lab](https://www.pymc-labs.io) 850 | 851 | * [CausalNex](https://github.com/quantumblacklabs/causalnex) A toolkit for causal reasoning with Bayesian Networks 852 | from Quantumblack. 853 | 854 | * Tetrad Project: Graphical Causal Models 855 | [url](http://www.phil.cmu.edu/tetrad/) 856 | 857 | ### Views 858 | 859 | * R Universe: [CRAN Task View Causal Inference](https://cran.r-project.org/web/views/CausalInference.html) 860 | 861 | ## Datasets 862 | 863 | * R causaldata 864 | Nick Huntington-Klein, Malcolm Barrett 865 | Example dataset from some of the Causality textbooks 866 | [github](https://github.com/NickCH-K/causaldata) | [cran](https://cran.r-project.org/web/packages/causaldata/index.html) 867 | 868 | * Distinguishing Cause from Effect Using Observational Data: Methods and Benchmarks. 869 | Joris M. Mooij, Jonas Peters, Dominik Janzing, Jakob Zscheischler, Bernhard Schölkopf; 870 | 17(32):1−102, (2016). [jmlr-paper](https://jmlr.org/papers/v17/14-518.html). 871 | Database with cause-effect pairs [url](https://webdav.tuebingen.mpg.de/cause-effect/). 872 | 873 | ## MOOCs 874 | 875 | * Mixtape sessions [url](https://www.mixtapesessions.io) 876 | Free Course Material 877 | 878 | * A Crash Course in Causality: Inferring Causal Effects from Observational Data. 879 | [url](https://www.coursera.org/learn/crash-course-in-causality) 880 | 881 | * Measuring Causal Effects in the Social Sciences. 882 | [url](https://www.coursera.org/learn/causal-effects) 883 | 884 | * Probabilistic Graphical Models Specialization. 885 | Daphne Koller. 886 | [url](https://www.coursera.org/specializations/probabilistic-graphical-models) 887 | 888 | ## Blog Posts 889 | 890 | * Resolving disputes between J. Pearl and D. Rubin on causal inference. 891 | Gelman (2009) [url](https://statmodeling.stat.columbia.edu/2009/07/05/disputes_about/). 892 | Follow ups [1](https://statmodeling.stat.columbia.edu/2009/07/07/more_on_pearls/) 893 | [2](https://statmodeling.stat.columbia.edu/2009/07/09/more_on_pearlru/) 894 | [3](https://statmodeling.stat.columbia.edu/2009/07/07/philip_dawids_t/) 895 | 896 | * 05/2022 Statisticians vs. Causal Inference [url](https://statmodeling.stat.columbia.edu/2022/05/14/causal-is-what-we-say-when-we-dont-know-what-were-doing/#comment-2053584) | Pearl's response [url](http://causality.cs.ucla.edu/blog/index.php/2022/05/17/what-statisticians-mean-by-causal-inference-is-gelmans-blog-representative/) 897 | 898 | * Judea Pearl on Potential Outcomes. 899 | [url](http://causality.cs.ucla.edu/blog/index.php/2012/12/03/judea-pearl-on-potential-outcomes/) 900 | 901 | * Judea Pearl, model vs. model-free : 902 | "...no “causal interpretation” is needed for parameters that are intrinsically causal..." 903 | [url](http://causality.cs.ucla.edu/blog/index.php/2017/02/22/winter-2017-greeting-from-ucla-causality-blog/). 904 | 905 | ## Quotes 906 | 907 | Including interviews. 908 | 909 | * Causal Inference: History, Perspectives, Adventures, and Unification (An Interview with Judea Pearl). 910 | Observational Studies, 8, 2, 2022, p1-14. 911 | [url](https://muse.jhu.edu/article/867087) | [pdf](https://muse.jhu.edu/pub/56/article/867087/pdf). 912 | 913 | * Interview with Don Rubin. 914 | Observational Studies (2022). 915 | [url](https://muse.jhu.edu/article/867089). 916 | 917 | * A Conversation with A. Philip Dawid (2023) 918 | [arXiv](https://arxiv.org/abs/2312.00632) 919 | 920 | 921 | `Prediction and Causation` 922 | 923 | * " Actually correlation lets you make predictions 924 | in many cases, assuming you're making prediction 925 | about the world as reflected in your data. 926 | For example, correlations between photos and 927 | their labels allows you usually to make predictions 928 | about new photos. 929 | 930 | The problem gets difficult if you want to predict the 931 | effects of actions taken in a different manner from 932 | that which exists in your data. For example, if you 933 | want to consider different treatment strategies for 934 | cancer using data from past cancer patients, only 935 | correlations will usually not suffice as you're 936 | trying to predict counterfactual that might not exist in your data." 937 | Uri Shalit, Technion (Forum Communication 01/2018) 938 | 939 | `Rubin vs. Pearl` 940 | 941 | * " Rubin and Pearl are kind of "academic enemies". 942 | Though neither completely dismisses the other, 943 | they both make snide remarks about the other's work. 944 | Pearl shows in his book exactly how Neyman-Rubin 945 | potential outcomes can be derived from causal graphs. 946 | As far as I know Rubin never really makes an 947 | attempt to address Pearl's ideas directly. 948 | However, Rubin, being a statistician, made 949 | significant contributions to the practice of real-world 950 | causal inference, which go beyond Pearl's interests. 951 | Jamie Robins also made seminal contributions to this subject. 952 | You can read some of the debate on Andrew Gelman's blog 953 | [here](http://andrewgelman.com/2009/07/05/disputes_about/) 954 | Pearl writes in the comment section and in that blog 955 | post there are links to follow up posts. " 956 | Uri Shalit, Technion (Forum Communication 01/2018) 957 | 958 | `On the Pearl's Philosopy` 959 | 960 | * "..while Pearl's work is foundational, and its importance 961 | cannot be overstated, his published work is often 962 | insufficient in addressing the real-world problems of 963 | many data scientists. The reason is that Pearl is mostly 964 | concerned with the problem of identification, i.e. which data 965 | generating processes allow us to infer causation from observed data. 966 | He is less concerned with the statistical problem of actually 967 | inferring these purported causal relationships from data. 968 | This is especially true if the data is high-dimensional 969 | or noisy (Pearl usually considers a few binary or Gaussian variables)." 970 | Uri Shalit, Technion (Forum Communication 01/2018) 971 | 972 | ## Video Lectures 973 | including discussions 974 | 975 | * 2022 Causal Data Science. 976 | Elias Bareinboim (@ 1st Workshop on Interactive Causal Learning) [youtube](https://www.youtube.com/watch?v=6lWB_JmWY8g). 977 | * 2022 The Arrow of Time in Causal Networks 978 | Sean Carroll, Simons Intitute Talk [youtube](https://www.youtube.com/watch?v=6slug9rjaIQ) 979 | * Judea Pearl: Causal Reasoning, Counterfactuals, and the Path to AGI | Lex Fridman Podcast #56. 980 | Dec 11, 2019 [yt](https://www.youtube.com/watch?v=pEBI0vF45ic). 981 | * 2016 The Mathematics of Causal Inference: with Reflections on Machine Learning. 982 | Judea Pearl, Microsoft Research. [youtube](https://www.youtube.com/watch?v=bcRl7sXR1hE). 983 | * 2015 Toward Causal Machine Learning. 984 | Bernhard Schölkopf, [youtube](https://www.youtube.com/watch?v=ooeRlw3U2zU) 985 | * 2012 Causal Inference Conference Featuring Judea Pearl [youtube](https://www.youtube.com/watch?v=j4JOR-PQuqQ&t=185s). 986 | Judea Pearl's "The Mathematics of Causal Inference: Use it or Lose It" 987 | 988 | ## Contemporary Academics 989 | 990 | * [Judea Pearl](http://bayes.cs.ucla.edu/jp_home.html). 991 | * [Clark Glymour](https://www.cmu.edu/dietrich/philosophy/people/emeritus/glymour.html). 992 | * [Peter Spirtes](https://www.cmu.edu/dietrich/philosophy/people/faculty/spirtes.html) 993 | * [Richard Scheines](https://www.cmu.edu/dietrich/philosophy/people/faculty/scheines.html) 994 | * [Elias Bareinboim](https://causalai.net). 995 | * [Donald B. Rubin](https://statistics.fas.harvard.edu/people/donald-b-rubin). 996 | * [Joshua Angrist](https://economics.mit.edu/faculty/angrist). 997 | * [Guido W. Imbens](https://www.gsb.stanford.edu/faculty-research/faculty/guido-w-imbens). 998 | * [Andrew Gelman](http://www.stat.columbia.edu/~gelman/). 999 | * [David Card](https://davidcard.berkeley.edu). 1000 | * [Bernhard Schölkopf](https://is.mpg.de/~bs). 1001 | * [Joris M. Mooij](https://staff.fnwi.uva.nl/j.m.mooij/). 1002 | * [Dominik Janzing](https://janzing.github.io). 1003 | * [Uri Shalit](https://dds.technion.ac.il/academicstaff/uri-shalit/). 1004 | 1005 | # Communities Conferences 1006 | 1007 | * Causal RL [RLC Workshop 2025](https://sites.google.com/uci.edu/crlw2025/) 1008 | * [Causal Science](https://www.causalscience.org) 1009 | * Are Large Language Models Simply Causal Parrots? 1010 | Annual AAAI Conference on Artificial Intelligence 2024 [url](https://llmcp.cause-lab.net/schedule-llmcp) 1011 | * Counterfactual reasoning: From minds to machines to practical applications. 1012 | ICML Workshop [url](https://sites.google.com/view/counterfactuals-icml/home) (2023) 1013 | * Neurips 2023: Causal Representation Learning (CRL) [url](https://neurips.cc/virtual/2023/workshop/66497) 1014 | 1015 | -------------------------------------------------------------------------------- /nuggets/LN1.md: -------------------------------------------------------------------------------- 1 | # LN 1: weighted Directed Acyclic Graph (wDAG) 2 | 3 | Definition (wDAG): A weighted Directed Acyclic Graph (wDAG) $\mathscr{G_{c}}$ is defined as set of ordered triplets of weights and connected random variables, such that, $k$th triplet $(w_{k}, x_{i}, x_{j})$ where by $w_{k} \in \mathbb{R}$ is the weight, an effect size, between two variates that $x_{i}$ effects $x_{j}$. There are constraints : 4 | 5 | (i) No cyclic effects can be defined, necessarily $x_{i}$ can not be equal to $ x_{j}$. 6 | 7 | (ii) If there is a definition, $(w_{k}, x_{i}, x_{j})$ the reverse can't be defined, i.e., so that $(w_{k}, x_{j}, x_{i})$ does not exist. 8 | 9 | (iii) No two causal effects sizes can't be exactly equal, $w_{k}$ can not be equal to $w_{l}$, from the same causal variable, meaning no simultaneous events caused by the same random variable. This prevents ambiguity of ordering and random tie-breaks are unnatural. 10 | 11 | ## Further reading and notes 12 | 13 | * Pearl, Glymour and Jewell. 14 | Causal Inference in Statistics: A Primer (2016). 15 | [amzn](https://www.amzn.com/dp/1119186846). 16 | [Ch4-pdf](http://web.cs.ucla.edu/~kaoru/primer-ch4.pdf). 17 | [tweet-solution-manual](https://twitter.com/yudapearl/status/1484023795811696642). 18 | 19 | * Constraint (iii) may not be applied, then ties should be break randomly, in the case of ordered events. 20 | 21 | [Cite-me](../CITEME.md) 22 | -------------------------------------------------------------------------------- /nuggets/looper_nuggets.md: -------------------------------------------------------------------------------- 1 | # Looper Nugget 2 | 3 | * [LN1](LN1.md): weighted-Directed Acyclic Graph (wDAG) 4 | 5 | ## Citing Nuggets 6 | 7 | [Cite-me](../CITEME.md) 8 | -------------------------------------------------------------------------------- /src/README.md: -------------------------------------------------------------------------------- 1 | src 2 | --------------------------------------------------------------------------------