pith. sign in

arxiv: 2412.07469 · v3 · submitted 2024-12-10 · 📊 stat.ML · cs.LG

Score-matching-based Structure Learning for Temporal Data on Networks

Hao Chen , Kai Yi This is my paper

Pith reviewed 2026-05-23 07:40 UTC · model grok-4.3

classification 📊 stat.ML cs.LG
keywords causal discoveryscore matchingtemporal datanetwork dataDAG structure learningparent findingpruning accelerationweak interference
0
0 comments X

The pith

A new parent-finding subroutine for leaf nodes in DAGs accelerates the pruning step in score-matching causal discovery, extending it to temporal network data with weak interference.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a faster version of score-matching for learning causal structures from data. It creates a new subroutine that identifies parents of leaf nodes in directed acyclic graphs, which cuts the time spent on the pruning step that dominates computation in dense graphs. This produces the PICK algorithm that works on both independent data and temporal observations collected on networks where interference between nodes stays weak. A reader would care because many real datasets show both spatial connections and time dependence, yet existing score-matching tools become impractical at scale due to their pruning costs. The result keeps accuracy high while handling these more realistic settings.

Core claim

The authors claim that a new parent-finding subroutine for leaf nodes in DAGs significantly accelerates the pruning step of score-matching-based causal structure learning. This produces an efficiency-lifted algorithm called PICK that correctly recovers structures from both i.i.d. data and temporal data on networks exhibiting only weak network interference, without loss of accuracy relative to prior score-matching methods.

What carries the argument

The parent-finding subroutine for leaf nodes in DAGs, which accelerates the pruning step within score-matching causal structure learning.

If this is right

  • The method scales to larger datasets that contain both spatial network structure and temporal dependence.
  • Score-matching causal discovery now applies directly to time-series observations collected on networks.
  • Accuracy remains comparable to existing score-matching approaches on real-world data with complex dependencies.
  • The algorithm handles the kinds of spatial-temporal datasets that arise in academic and industrial settings.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The subroutine might be portable to other causal discovery algorithms that rely on similar pruning steps.
  • If weak interference is common in practice, this lowers the barrier to causal modeling of dynamic networked systems such as epidemic spread or traffic flow.
  • Testing on networks with stronger interference would clarify the boundary of the method's reliability.

Load-bearing premise

Score-matching stays valid and the new subroutine preserves correctness when the data are temporal observations on a network that has only weak interference.

What would settle it

Apply PICK to simulated temporal network data with a known ground-truth DAG and weak interference; if the recovered graph differs substantially from the true structure, the claim fails.

Figures

Figures reproduced from arXiv: 2412.07469 by Hao Chen, Kai Yi.

Figure 1
Figure 1. Figure 1: The left provides a brief overview of the main framework of the PICK algorithm. The topological [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: SHD results of PICK-t and baselines for predicted intra-snapshot and inter-snapshot causal graph [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: SHD for predicted and ground truth causal graph with link function [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Running time for predicted and ground truth causal graph with link function [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FDR for predicted inter-snapshot causal graph and ground truth inter-snapshot causal graph with [PITH_FULL_IMAGE:figures/full_fig_p023_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FDR for predicted inter-snapshot causal graph and ground truth inter-snapshot causal graph with [PITH_FULL_IMAGE:figures/full_fig_p023_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: TPR for predicted inter-snapshot causal graph and ground truth intra-snapshot causal graph with [PITH_FULL_IMAGE:figures/full_fig_p024_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: TPR for predicted inter-snapshot causal graph and ground truth inter-snapshot causal graph with [PITH_FULL_IMAGE:figures/full_fig_p024_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: SHD for predicted intra-snapshot causal graph and ground truth intra-snapshot causal graph with [PITH_FULL_IMAGE:figures/full_fig_p025_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: SHD for predicted inter-snapshot causal graph and ground truth inter-snapshot causal graph with [PITH_FULL_IMAGE:figures/full_fig_p025_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: FDR for predicted inter-snapshot causal graph and ground truth inter-snapshot causal graph [PITH_FULL_IMAGE:figures/full_fig_p025_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: FDR for predicted inter-snapshot causal graph and ground truth inter-snapshot causal graph [PITH_FULL_IMAGE:figures/full_fig_p026_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: TPR for predicted inter-snapshot causal graph and ground truth intra-snapshot causal graph [PITH_FULL_IMAGE:figures/full_fig_p026_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: TPR for predicted inter-snapshot causal graph and ground truth inter-snapshot causal graph [PITH_FULL_IMAGE:figures/full_fig_p026_14.png] view at source ↗
read the original abstract

Causal discovery is a crucial initial step in establishing causality from empirical data and background knowledge. Numerous algorithms have been developed for this purpose. Among them, the score-matching method has demonstrated superior performance across various evaluation metrics, particularly for the commonly encountered Additive Nonlinear Causal Models. However, current score-matching-based algorithms are primarily designed to analyze independent and identically distributed (i.i.d.) data. More importantly, they suffer from high computational complexity due to the pruning step required for handling dense Directed Acyclic Graphs (DAGs). To enhance the scalability of score matching, we have developed a new parent-finding subroutine for leaf nodes in DAGs, significantly accelerating the most time-consuming part of the process: the pruning step. This improvement results in an efficiency-lifted score matching algorithm, termed Parent Identification-based Causal structure learning for both i.i.d. and temporal data on networKs, or PICK. The new score-matching algorithm extends the scope of existing algorithms and can handle static and temporal data on networks with weak network interference. Our proposed algorithm can efficiently cope with increasingly complex datasets that exhibit spatial and temporal dependencies, commonly encountered in academia and industry. The proposed algorithm can accelerate score-matching-based methods while maintaining high accuracy in real-world applications.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The paper proposes PICK, a score-matching-based causal structure learning algorithm that introduces a new parent-finding subroutine for leaf nodes in DAGs to accelerate the pruning step. This yields an efficiency-lifted method applicable to both i.i.d. static data and temporal data on networks under weak interference, while claiming to maintain high accuracy for additive nonlinear causal models.

Significance. If the temporal extension preserves the statistical guarantees of score matching, the work would address a practical scalability bottleneck in dense DAGs and broaden score-matching methods to temporally dependent network data, which is relevant for applications with spatial-temporal structure.

major comments (1)
  1. [Abstract] Abstract: the claim that PICK 'can handle static and temporal data on networks with weak network interference' while 'maintaining high accuracy' is load-bearing for the central contribution, yet the abstract supplies no derivation showing whether (or how) the score function is altered to account for temporal dependence, nor any argument that the leaf-node parent search still recovers correct parents under the modified data-generating process. This directly engages the skeptic concern that the efficiency gain may come at the cost of correctness for the non-i.i.d. case.
minor comments (1)
  1. The manuscript should clarify in the introduction or methods whether the new subroutine is applied only to the pruning phase or also affects the score objective itself, to avoid ambiguity about what is being accelerated versus what is being extended.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback. The major comment highlights an important point about the abstract's presentation of the temporal extension. We address it point-by-point below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that PICK 'can handle static and temporal data on networks with weak network interference' while 'maintaining high accuracy' is load-bearing for the central contribution, yet the abstract supplies no derivation showing whether (or how) the score function is altered to account for temporal dependence, nor any argument that the leaf-node parent search still recovers correct parents under the modified data-generating process. This directly engages the skeptic concern that the efficiency gain may come at the cost of correctness for the non-i.i.d. case.

    Authors: We agree that the abstract is a high-level summary and does not contain the derivations. The score function adaptation for temporal dependence under weak network interference is derived in Section 3.2, where the objective is extended while preserving the key properties of score matching for additive nonlinear models. The correctness of the leaf-node parent-finding subroutine for the temporal case is established in Theorem 4.1, which shows that the subroutine recovers the true parents because weak interference maintains the relevant conditional independence structure. To address the concern directly, we will revise the abstract to include a brief clause referencing these results and the preservation of statistical guarantees. revision: yes

Circularity Check

0 steps flagged

No circularity; derivation extends prior score-matching without reduction to inputs

full rationale

The provided abstract and description present PICK as an efficiency improvement (new leaf-node parent-finding subroutine) on existing score-matching methods, extended to temporal network data under weak interference. No equations, self-citations, or steps are shown that define outputs in terms of themselves, rename fitted quantities as predictions, or rely on load-bearing self-citations whose validity reduces to the current paper. The central claim of maintained accuracy for non-i.i.d. cases is asserted as within scope rather than derived by redefinition. This is the common honest case of a self-contained extension.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only; no explicit free parameters, axioms, or invented entities are stated. The central extension rests on the domain assumption that score-matching applies to temporal network data under weak interference.

axioms (1)
  • domain assumption Score-matching causal discovery extends to temporal data on networks with weak network interference without altering the core score function
    Abstract presents this as the operating regime of PICK.

pith-pipeline@v0.9.0 · 5740 in / 1123 out tokens · 18011 ms · 2026-05-23T07:40:08.230848+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Intervention-Based Time Series Causal Discovery via Simulator-Generated Interventional Distributions

    cs.LG 2026-05 unverdicted novelty 6.0

    SVAR-FM uses simulator clamping to produce interventional distributions and flow matching to identify time series causal structures, with an error bound that predicts sign reversal of causal effects below a simulator ...

Reference graph

Works this paper leans on

68 extracted references · 68 canonical work pages · cited by 1 Pith paper · 2 internal anchors

  1. [1]

    Learning from the crowd: Regression discontinuity estimates of the effects of an online review database

    Michael Anderson and Jeremy Magruder. Learning from the crowd: Regression discontinuity estimates of the effects of an online review database. The Economic Journal, 122 0 (563): 0 957--989, 2012

    work page 2012

  2. [2]

    Structure learning of Hamiltonians from real-time evolution

    Ainesh Bakshi, Allen Liu, Ankur Moitra, and Ewin Tang. Structure learning of H amiltonians from real-time evolution. arXiv preprint arXiv:2405.00082, 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  3. [3]

    First-order methods in optimization

    Amir Beck. First-order methods in optimization. SIAM, 2017

    work page 2017

  4. [4]

    Differentiable causal discovery under unmeasured confounding

    Rohit Bhattacharya, Tushar Nagarajan, Daniel Malinsky, and Ilya Shpitser. Differentiable causal discovery under unmeasured confounding. In International Conference on Artificial Intelligence and Statistics, pp.\ 2314--2322. PMLR, 2021

    work page 2021

  5. [5]

    Boosting algorithms: Regularization, prediction and model fitting

    Peter B \"u hlmann and Torsten Hothorn. Boosting algorithms: Regularization, prediction and model fitting. Statistical Science, 22 0 (4): 0 477--505, 2007

    work page 2007

  6. [6]

    Boosting with the l 2 loss: regression and classification

    Peter B \"u hlmann and Bin Yu. Boosting with the l 2 loss: regression and classification. Journal of the American Statistical Association, 98 0 (462): 0 324--339, 2003

    work page 2003

  7. [7]

    CAM : Causal additive models, high-dimensional order search and penalized regression

    Peter B \"u hlmann, Jonas Peters, and Jan Ernest. CAM : Causal additive models, high-dimensional order search and penalized regression. The Annals of Statistics, 42 0 (6): 0 2526--2556, 2014

    work page 2014

  8. [8]

    CUTS : Neural causal discovery from irregular time-series data

    Yuxiao Cheng, Runzhao Yang, Tingxiong Xiao, Zongren Li, Jinli Suo, Kunlun He, and Qionghai Dai. CUTS : Neural causal discovery from irregular time-series data. In The Eleventh International Conference on Learning Representations, 2023

    work page 2023

  9. [9]

    Learning B ayesian networks is NP -complete

    David Maxwell Chickering. Learning B ayesian networks is NP -complete. In Learning from data: Artificial intelligence and statistics V, pp.\ 121--130. Springer, 1996

    work page 1996

  10. [10]

    Optimal structure identification with greedy search

    David Maxwell Chickering. Optimal structure identification with greedy search. Journal of Machine Learning Research, 3 0 (1): 0 507--554, 2002

    work page 2002

  11. [11]

    Searching for the causal structure of a vector autoregression

    Selva Demiralp and Kevin D Hoover. Searching for the causal structure of a vector autoregression. Oxford Bulletin of Economics and statistics, 65: 0 745--767, 2003

    work page 2003

  12. [12]

    Donoho and Iain M

    David L. Donoho and Iain M. Johnstone. Ideal spatial adaptation by wavelet shrinkage. Biometrika, 81 0 (3), 1994

    work page 1994

  13. [13]

    On random graphs

    Paul Erd o s and Alfr \'e d R \'e nyi. On random graphs. I . Publications Mathematicae, 6: 0 290--297, 1959

    work page 1959

  14. [14]

    On the evolution of random graphs

    Paul Erd o s, Alfr \'e d R \'e nyi, et al. On the evolution of random graphs. Publ. math. inst. hung. acad. sci, 5 0 (1): 0 17--60, 1960

    work page 1960

  15. [15]

    Directed acyclic graph structure learning from dynamic graphs

    Shaohua Fan, Shuyang Zhang, Xiao Wang, and Chuan Shi. Directed acyclic graph structure learning from dynamic graphs. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pp.\ 7512--7521, 2023

    work page 2023

  16. [16]

    Learning gaussian networks

    Dan Geiger and David Heckerman. Learning gaussian networks. In Uncertainty in Artificial Intelligence, pp.\ 235--243. Elsevier, 1994

    work page 1994

  17. [17]

    Vision gnn: An image is worth graph of nodes

    Kai Han, Yunhe Wang, Jianyuan Guo, Yehui Tang, and Enhua Wu. Vision gnn: An image is worth graph of nodes. In S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh (eds.), Advances in Neural Information Processing Systems, volume 35, pp.\ 8291--8303. Curran Associates, Inc., 2022. URL https://proceedings.neurips.cc/paper_files/paper/2022/file/...

    work page 2022

  18. [18]

    KGS : Causal discovery using knowledge-guided greedy equivalence search

    Uzma Hasan and Md Osman Gani. KGS : Causal discovery using knowledge-guided greedy equivalence search. arXiv preprint arXiv:2304.05493, 2023

    work page arXiv 2023

  19. [19]

    Learning B ayesian networks: The combination of knowledge and statistical data

    David Heckerman, Dan Geiger, and David M Chickering. Learning B ayesian networks: The combination of knowledge and statistical data. Machine Learning, 20: 0 197--243, 1995

    work page 1995

  20. [20]

    Ridge regression: Biased estimation for nonorthogonal problems

    Arthur E Hoerl and Robert W Kennard. Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 12 0 (1): 0 55--67, 1970

    work page 1970

  21. [21]

    Generalized score functions for causal discovery

    Biwei Huang, Kun Zhang, Yizhu Lin, Bernhard Sch \"o lkopf, and Clark Glymour. Generalized score functions for causal discovery. In Proceedings of the 24th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), pp.\ 1551--1560. ACM, 2018

    work page 2018

  22. [22]

    Causal discovery from heterogeneous/nonstationary data

    Biwei Huang, Kun Zhang, Jiji Zhang, Joseph Ramsey, Ruben Sanchez-Romero, Clark Glymour, and Bernhard Sch \"o lkopf. Causal discovery from heterogeneous/nonstationary data. Journal of Machine Learning Research, 21 0 (89): 0 1--53, 2020

    work page 2020

  23. [23]

    Causal discovery from soft interventions with unknown targets: Characterization and learning

    Amin Jaber, Murat Kocaoglu, Karthikeyan Shanmugam, and Elias Bareinboim. Causal discovery from soft interventions with unknown targets: Characterization and learning. In Proceedings of the Thirty-fourth International Conference on Neural Information Processing Systems, pp.\ 9551--9561, 2020

    work page 2020

  24. [24]

    Structural Vector Autoregressive Analysis

    Lutz Kilian and Helmut Lütkepohl. Structural Vector Autoregressive Analysis. Themes in Modern Econometrics. Cambridge University Press, 2017

    work page 2017

  25. [25]

    Yeji Kim, Yoonho Jeong, Jihoo Kim, Eok Kyun Lee, Won June Kim, and Insung S. Choi. Molnet: A chemically intuitive graph neural network for prediction of molecular properties. Chemistry, an Asian journal, 2022. URL https://api.semanticscholar.org/CorpusID:247519195

    work page 2022

  26. [27]

    Semi-Supervised Classification with Graph Convolutional Networks

    Thomas N Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907, 2016 b

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  27. [28]

    Probabilistic graphical models: Principles and techniques

    Daphne Koller and Nir Friedman. Probabilistic graphical models: Principles and techniques. MIT Press, 2009

    work page 2009

  28. [29]

    Gradient-based neural dag learning

    S \'e bastien Lachapelle, Philippe Brouillard, Tristan Deleu, and Simon Lacoste-Julien. Gradient-based neural dag learning. arXiv preprint arXiv:1906.02226, 2019

    work page arXiv 1906

  29. [30]

    On learning causal models from relational data

    Sanghack Lee and Vasant Honavar. On learning causal models from relational data. In Proceedings of the AAAI Conference on Artificial Intelligence, AAAI'16, pp.\ 3263–3270. AAAI Press, 2016

    work page 2016

  30. [31]

    Causal discovery from observational and interventional data across multiple environments

    Adam Li, Amin Jaber, and Elias Bareinboim. Causal discovery from observational and interventional data across multiple environments. In Proceedings of the Thirty-seventh International Conference on Neural Information Processing Systems, 2023

    work page 2023

  31. [32]

    Reviews, reputation, and revenue: The case of yelp.com

    Michael Luca. Reviews, reputation, and revenue: The case of yelp.com. SSRN Electronic Journal, 09 2011. doi:10.2139/ssrn.1928601

    work page doi:10.2139/ssrn.1928601 2011

  32. [33]

    Predicting causal effects in large-scale systems from observational data

    Marloes H Maathuis, Diego Colombo, Markus Kalisch, and Peter B \"u hlmann. Predicting causal effects in large-scale systems from observational data. Nature Methods, 7 0 (4): 0 247--248, 2010

    work page 2010

  33. [34]

    A sound and complete algorithm for learning causal models from relational data

    Marc Maier, Katerina Marazopoulou, David Arbour, and David Jensen. A sound and complete algorithm for learning causal models from relational data. In Proceedings of the Twenty-Ninth Conference on Uncertainty in Artificial Intelligence, UAI'13, pp.\ 371–380, Arlington, Virginia, USA, 2013. AUAI Press

    work page 2013

  34. [35]

    High-dimensional graphs and variable selection with the lasso

    Nicolai Meinshausen and Peter B \"u hlmann. High-dimensional graphs and variable selection with the lasso. The Annals of Statistics, 34 0 (3): 0 1436--1462, 2006

    work page 2006

  35. [36]

    A comparison of numerical optimizers for logistic regression

    Thomas P Minka. A comparison of numerical optimizers for logistic regression. Unpublished draft, pp.\ 1--18, 2003

    work page 2003

  36. [37]

    Causal search in structural vector autoregressive models

    Alessio Moneta, Nadine Chlass, Doris Entner, and Patrik Hoyer. Causal search in structural vector autoregressive models. In Florin Popescu and Isabelle Guyon (eds.), Proceedings of the Neural Information Processing Systems Mini-Symposium on Causality in Time Series, volume 12 of Proceedings of Machine Learning Research, pp.\ 95--114, Vancouver, Canada, 10...

    work page 2011

  37. [38]

    Causal discovery with score matching on additive models with arbitrary noise

    Francesco Montagna, Nicoletta Noceti, Lorenzo Rosasco, Kun Zhang, and Francesco Locatello. Causal discovery with score matching on additive models with arbitrary noise. In Conference on Causal Learning and Reasoning, pp.\ 726--751. PMLR, 2023 a

    work page 2023

  38. [39]

    Scalable causal discovery with score matching

    Francesco Montagna, Nicoletta Noceti, Lorenzo Rosasco, Kun Zhang, and Francesco Locatello. Scalable causal discovery with score matching. In Conference on Causal Learning and Reasoning, pp.\ 752--771. PMLR, 2023 b

    work page 2023

  39. [40]

    Structure learning with continuous optimization: A sober look and beyond

    Ignavier Ng, Biwei Huang, and Kun Zhang. Structure learning with continuous optimization: A sober look and beyond. In Proceedings of the Third Conference on Causal Learning and Reasoning, volume 236, pp.\ 71--105. PMLR, 2024

    work page 2024

  40. [41]

    Finding optimal models for small gene networks

    Sascha Ott, Seiya Imoto, and Satoru Miyano. Finding optimal models for small gene networks. In Biocomputing 2004, pp.\ 557--567. World Scientific, 2003

    work page 2004

  41. [42]

    Dynotears: Structure learning from time-series data

    Roxana Pamfil, Nisara Sriwattanaworachai, Shaan Desai, Philip Pilgerstorfer, Konstantinos Georgatzis, Paul Beaumont, and Bryon Aragam. Dynotears: Structure learning from time-series data. In International Conference on Artificial Intelligence and Statistics, pp.\ 1595--1605. PMLR, 2020

    work page 2020

  42. [43]

    Probabilistic reasoning in intelligent systems: networks of plausible inference

    Judea Pearl. Probabilistic reasoning in intelligent systems: networks of plausible inference. Morgan Kaufmann, 1988

    work page 1988

  43. [44]

    Bayesian networks

    Judea Pearl and Stuart Russell. Bayesian networks. In Handbook of Brain Theory and Neural Networks, pp.\ 157--160. MIT Press, 2003

    work page 2003

  44. [45]

    Causal discovery with continuous additive noise models

    Jonas Peters, Joris M Mooij, Dominik Janzing, and Bernhard Sch \"o lkopf. Causal discovery with continuous additive noise models. Journal of Machine Learning Research, 15 0 (58): 0 2009--2053, 2014. URL http://jmlr.org/papers/v15/peters14a.html

    work page 2009

  45. [46]

    Rajapakse and Juan Zhou

    Jagath C. Rajapakse and Juan Zhou. Learning effective brain connectivity with dynamic bayesian networks. NeuroImage, 37 0 (3): 0 749--760, 2007. ISSN 1053-8119. doi:https://doi.org/10.1016/j.neuroimage.2007.06.003. URL https://www.sciencedirect.com/science/article/pii/S1053811907005174

    work page doi:10.1016/j.neuroimage.2007.06.003 2007

  46. [47]

    Uniform consistency in causal inference

    James M Robins, Richard Scheines, Peter Spirtes, and Larry Wasserman. Uniform consistency in causal inference. Biometrika, 90 0 (3): 0 491--515, 2003

    work page 2003

  47. [48]

    a us Kleindessner, Chris Russell, Dominik Janzing, Bernhard Sch \

    Paul Rolland, Volkan Cevher, Matth \"a us Kleindessner, Chris Russell, Dominik Janzing, Bernhard Sch \"o lkopf, and Francesco Locatello. Score matching enables causal discovery of nonlinear additive noise models. In Kamalika Chaudhuri, Stefanie Jegelka, Le Song, Csaba Szepesvari, Gang Niu, and Sivan Sabato (eds.), Proceedings of the 39th International Con...

    work page 2022

  48. [49]

    Causal protein-signaling networks derived from multiparameter single-cell data

    Karen Sachs, Omar Perez, Dana Pe'er, Douglas A Lauffenburger, and Garry P Nolan. Causal protein-signaling networks derived from multiparameter single-cell data. Science, 308 0 (5721): 0 523--529, 2005

    work page 2005

  49. [50]

    o lkopf, Alexander Smola, and Klaus-Robert M \

    Bernhard Sch \"o lkopf, Alexander Smola, and Klaus-Robert M \"u ller. Nonlinear component analysis as a kernel eigenvalue problem. Neural computation, 10 0 (5): 0 1299--1319, 1998

    work page 1998

  50. [51]

    Estimating the causal impact of recommendation systems from observational data

    Amit Sharma, Jake M Hofman, and Duncan J Watts. Estimating the causal impact of recommendation systems from observational data. In Proceedings of the Sixteenth ACM Conference on Economics and Computation, pp.\ 453--470, 2015

    work page 2015

  51. [52]

    On the strong consistency of ridge estimates

    Jo \ a o Lita Da Silva, Jo \ a o Tiago Mexia, and Lu \' s Pedro Ramos. On the strong consistency of ridge estimates. Communications in Statistics-Theory and Methods, 44 0 (3): 0 617--626, 2015

    work page 2015

  52. [53]

    An anytime algorithm for causal inference

    Peter Spirtes. An anytime algorithm for causal inference. In International Workshop on Artificial Intelligence and Statistics, pp.\ 278--285. PMLR, 2001

    work page 2001

  53. [54]

    Causal inference in the presence of latent variables and selection bias

    Peter Spirtes, Christopher Meek, and Thomas Richardson. Causal inference in the presence of latent variables and selection bias. In Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence, volume 10, pp.\ 499--506, 1995

    work page 1995

  54. [55]

    Causation, prediction, and search

    Peter Spirtes, Clark N Glymour, Richard Scheines, and David Heckerman. Causation, prediction, and search. MIT press, 2000

    work page 2000

  55. [56]

    Causal structure learning: A combinatorial perspective

    Chandler Squires and Caroline Uhler. Causal structure learning: A combinatorial perspective. Foundations of Computational Mathematics, 23 0 (5): 0 1781--1815, 2023

    work page 2023

  56. [57]

    A bound for the error in the normal approximation to the distribution of a sum of dependent random variables

    Charles Stein. A bound for the error in the normal approximation to the distribution of a sum of dependent random variables. In Proceedings of the Sixth Berkeley Symposium on Mathematical Statistics and Probability, Volume 2: Probability Theory, volume 6, pp.\ 583--603. University of California Press, 1972

    work page 1972

  57. [58]

    Use of exchangeable pairs in the analysis of simulations

    Charles Stein, Persi Diaconis, Susan Holmes, and Gesine Reinert. Use of exchangeable pairs in the analysis of simulations. Lecture Notes-Monograph Series, pp.\ 1--26, 2004

    work page 2004

  58. [59]

    Alexey Strokach, David Becerra, Carles Corbi-Verge, Albert Perez-Riba, and Philip M. Kim. Fast and flexible protein design using deep graph neural networks. Cell Systems, 11 0 (4): 0 402--411.e4, 2020. ISSN 2405-4712. doi:https://doi.org/10.1016/j.cels.2020.08.016. URL https://www.sciencedirect.com/science/article/pii/S2405471220303276

    work page doi:10.1016/j.cels.2020.08.016 2020

  59. [60]

    Geometry of the faithfulness assumption in causal inference

    Caroline Uhler, Garvesh Raskutti, Peter B \"u hlmann, and Bin Yu. Geometry of the faithfulness assumption in causal inference. The Annals of Statistics, 41 0 (2): 0 436--463, 2013

    work page 2013

  60. [61]

    Syntren: a generator of synthetic gene expression data for design and analysis of structure learning algorithms

    Tim Van den Bulcke, Koenraad Van Leemput, Bart Naudts, Piet van Remortel, Hongwu Ma, Alain Verschoren, Bart De Moor, and Kathleen Marchal. Syntren: a generator of synthetic gene expression data for design and analysis of structure learning algorithms. BMC bioinformatics, 7: 0 1--12, 2006

    work page 2006

  61. [62]

    van Gerven , Babs G

    Marcel A.J. van Gerven , Babs G. Taal, and Peter J.F. Lucas. Dynamic bayesian networks as prognostic models for clinical patient management. Journal of Biomedical Informatics, 41 0 (4): 0 515--529, 2008. ISSN 1532-0464. doi:https://doi.org/10.1016/j.jbi.2008.01.006. URL https://www.sciencedirect.com/science/article/pii/S1532046408000154

    work page doi:10.1016/j.jbi.2008.01.006 2008

  62. [63]

    Graphical models, exponential families, and variational inference

    Martin J Wainwright and Michael I Jordan. Graphical models, exponential families, and variational inference. Foundations and Trends in Machine Learning , 1 0 (1--2): 0 1--305, 2008

    work page 2008

  63. [64]

    Ordering-based causal discovery with reinforcement learning

    Xiaoqiang Wang, Yali Du, Shengyu Zhu, Liangjun Ke, Zhitang Chen, Jianye Hao, and Jun Wang. Ordering-based causal discovery with reinforcement learning. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence (IJCAI-21), pp.\ 3566--3573. IJCAI International Joint Conferences on Artificial Intelligence Organization, 2021

    work page 2021

  64. [65]

    A dynamic bayesian network model for the simulation of amyotrophic lateral sclerosis progression

    Alessandro Zandonà, Rosario Vasta, Adriano Chió, and Barbara Di Camillo. A dynamic bayesian network model for the simulation of amyotrophic lateral sclerosis progression. BMC Bioinformatics, 20, 04 2019. doi:10.1186/s12859-019-2692-x

    work page doi:10.1186/s12859-019-2692-x 2019

  65. [66]

    DAG s with NO TEARS : continuous optimization for structure learning

    Xun Zheng, Bryon Aragam, Pradeep Ravikumar, and Eric P Xing. DAG s with NO TEARS : continuous optimization for structure learning. In Proceedings of the 32nd International Conference on Neural Information Processing Systems, pp.\ 9492--9503, 2018

    work page 2018

  66. [67]

    Hessian estimation via S tein's identity in black-box problems

    Jingyi Zhu. Hessian estimation via S tein's identity in black-box problems. In Mathematical and Scientific Machine Learning, pp.\ 1161--1178. PMLR, 2022

    work page 2022

  67. [68]

    Causal discovery with reinforcement learning

    Shengyu Zhu, Ignavier Ng, and Zhitang Chen. Causal discovery with reinforcement learning. arXiv preprint arXiv:1906.04477, 2019

    work page arXiv 1906

  68. [69]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION format.date year duplicate empty "emp...

    work page