Algorithmic Aspects of Causal Inference 2022
2022-03-18..2022-03-23
Berkeley, United States
[Quantum Physics and Statistical Causal Models (Simons Institute workshop, 2022), Learning from Interventions 2022, Algorithmic Advances for Statistical Inference with Combinatorial Structure, Quantifying Uncertainty: Stochastic, Adversarial, and Beyond, Meet the Fellows Welcome Event, Information Theory, Learning and Big Data (Simons Institute, 2015)]
who it's for
causal inference researchers, algorithm designers for learning, machine learning researchers, statistical learning theorists, fairness and accountability researchers, reinforcement learning researchers, network and systems researchers
topics
restricted Boltzmann machines learning; Gaussian graphical model recovery; preconditioning for sparse regression; causal matrix completion methods; mixture model identification algorithms; infinite-horizon off-policy evaluation; stability and condition numbers; probabilistic circuits tractability; multicalibration and fairness; atomic-intervention causal discovery; invariant causal experiment selection; kernel methods for causal inference
| title | author | length |
|---|---|---|
| Neural Networks And Spurious Correlations | Aditi Raghunathan | 40m |
Neural nets often latch onto easy but unstable cues, producing high worst-group error when spurious attributes (like background color) are easier to learn than… | ||
| Learning Restricted Boltzman Machines | Ankur Moitra | 44m |
Proves bounded‑degree RBMs exactly capture order‑T binary Markov random fields, giving a sharp representational characterization; shows improper learning of… | ||
| Learning Ill-Conditioned Gaussian Graphical Models | Raghu Meka | 32m |
Claims first fixed polynomial-time algorithms that learn attractive GGMs and walk-summable GGMs using logarithmic samples without condition-number assumptions… | ||
| Preconditioning In Sparse Linear Regression Using Graphical Structure | Frederic Koehler | 30m |
Shows a concrete claim: when covariates form a low treewidth Gaussian graphical model there is an explicit change-of-basis (a preconditioner) that preserves… | ||
| Causal Matrix Completion | Anish Agarwal | 29m |
Provides a causal model (potential outcomes) for matrix completion under MNAR, shows entries can be identified even when observation probability is zero, and… | ||
| Parameter Estimation For Undirected Graphical Models With Hard Constraints | Kavita Ramanan | 35m |
Parameter estimation for hard-constraint undirected graphical models uses the maximum pseudo-likelihood estimator, which for the hard-core model reduces to the… | ||
| Efficient Distance Estimation And Causal Inference For Discrete Models | Arnab Bhattacharyya | 43m |
Proves that TV computation is #P‑hard even for product distributions and that deciding TV zero versus nonzero becomes NP‑hard once node degree ≥2; gives an… | ||
| Identifying Mixture Models | Leonard J. Schulman | 40m |
Algorithm solves source identification for a mixture of k product distributions on n binary bits (k-MixProd), recovering the k-column parameter matrix from… | ||
| Identifying Mixtures Of Bayesian Network Distributions | Yuval Rabani | 43m |
Gives the first algorithm that identifies finite mixtures of Bayesian-network distributions beyond empty graphs by reducing to mixed PROD and then to mixed… | ||
| Confounding-Robust Policy Evaluation In Infinite-Horizon Reinforcement Learning | Angela Zhou | 28m |
Claims a method to compute sharp bounds on infinite-horizon off‑policy values under unobserved time‑step confounding using a marginal sensitivity model… | ||
| Stability Of Causal Identification From The Perspective Of Condition Numbers | Piyush Srivastava | 29m |
The work studies numerical and model robustness of causal identification by treating the symbolic output of the ID algorithm as a function of the observed… | ||
| Causalsim: Trace-Driven Simulation For Network Protocols | Devavrat Shah | 31m |
CausalSim is a trace-driven counterfactual simulator that learns a causal network model from an initial RCT (PUFFER) to remove policy bias, maps counterfactual… | ||
| Tractable Probabilistic Circuits | Guy Van den Broeck | 37m |
Introduces tractable probabilistic circuits as sum‑product DAGs learned via Chow‑Liu → hidden tree → circuit and trained with mini‑batch EM on GPUs; achieves… | ||
| A Multi-Group Approach To Algorithmic Fairness | Guy N. Rothblum | 50m |
Claims that multi-calibration yields predictors that are calibrated simultaneously over very rich, even exponential, families of overlapping subgroups and that… | ||
| Decision-Making Under Miscalibration | Gal Yona | 40m |
Formalizes a distribution-free decision rule that picks the threshold minimizing worst-case regret over all predictors with at most anticipated miscalibration… | ||
| Multicalibration, Universal Adaptability and Causality | Omer Reingold | 45m |
Shows a concrete reduction from adapting a study to many unseen targets to enforcing multicalibration: under conditional independence (Y ⟂ Z | X) and mild… | ||
| Algorithmic Fairness From The Lens Of Causality And Information Theory | Sanghamitra Dutta | 37m |
Proposes a concrete measure of non-exempt disparity that separates bias explainable by chosen critical features from bias that is not, using causality plus… | ||
| Orthogonal Statistical Learning | Vasilis Syrgkanis | 45m |
Provides finite-sample guarantees that make nuisance estimation a second-order error source by using a two-stage sample-splitting meta-algorithm plus Neyman… | ||
| The Approximate Implication Problem For Probabilistic Graphical Models | Batya Kenig | 48m |
Defines approximate implication as a linear inequality between mutual informations and asks when exact implications relax to such bounds, proving general… | ||
| Challenges For Causal Inference On Digital Platforms | Moritz Hardt | 48m |
A randomized audit uses 1% control / 4% treatment users on Twitter (April–Aug 2020) to define an amplification ratio and measures millions of political tweets… | ||
| Active Invariant Causal Prediction: Experiment Selection Through Stability | Christina Heinze-Deml | 40m |
Active Invariant Causal Prediction (called Active ICP) selects interventions to reveal direct parents of Y by testing which conditioning sets remain invariant… | ||
| Simple Yet Efficient Estimators For Network Causal Inference... | Christina Yu | 33m |
Shows that under a heterogeneous linear outcomes model the total treatment effect reduces to a population mean of per-node “influences,” so a plain Bernoulli… | ||
| Collaborative Causal Discovery With Atomic Interventions | Shiva Kasiviswanathan | 31m |
Defines a collaborative model that recovers multiple causal graphs (as Maximal Ancestral Graphs (MAGs)) when entities’ MAGs satisfy an α–β clustering… | ||
| New Approaches To Learning Nonparametric (Latent) Causal Graphical Models | Bryon Aragam | 32m |
Proves that under a subset condition on which latents connect to which observables and mild nondegeneracy, the bipartite graph and discrete latent distribution… | ||
| Kernel Methods For Causal Inference | Rahul Singh | 30m |
Proposes a practical method that turns long‑term causal questions into three matrix computations: use kernel ridge regressions to learn a surrogate regression… | ||
| nothing matches | ||