• Corpus ID: 238226888

Causal Matrix Completion

  title={Causal Matrix Completion},
  author={Anish Agarwal and Munther A. Dahleh and Devavrat Shah and Dennis Shen},
Matrix completion is the study of recovering an underlying matrix from a sparse subset of noisy observations. Traditionally, it is assumed that the entries of the matrix are “missing completely at random” (MCAR), i.e., each entry is revealed at random, independent of everything else, with uniform probability. This is likely unrealistic due to the presence of “latent confounders”, i.e., unobserved factors that determine both the entries of the underlying matrix and the missingness pattern in the… 

Figures and Tables from this paper

Truncated Matrix Completion - An Empirical Study

Through a series of experiments, this paper studies and compares the performance of various LRMC algorithms that were originally successful for data-independent sampling patterns and considers various settings where the sampling mask is dependent on the underlying data values.

Causal Inference for Social and Engineering Systems

  • A. Agarwal
  • Computer Science
    ACM SIGMETRICS Performance Evaluation Review
  • 2022
This work explores how to answer counterfactual questions using observational data-which is increasingly available due to digitization and pervasive sensors-and/or very limited experimental data, and discusses connections between matrix/tensor completion and time series analysis and reinforcement learning.

CausalSim: Unbiased Trace-Driven Network Simulation

Estimation of CausalSim on both real and synthetic datasets and two use cases shows that it provides accurate counterfactual predictions, reducing prediction error by 44% and 53% on average compared to expert-designed and supervised learning baselines.

Optimal Recovery for Causal Inference

It is crucial to successfully quantify causal effects of a policy intervention to determine whether the policy achieved the desired outcomes. We present a deterministic approach to a classical method

Synthetic Blip Effects: Generalizing Synthetic Controls for the Dynamic Treatment Regime

We propose a generalization of the synthetic control and synthetic interventions methodology to the dynamic treatment regime. We consider the estimation of unit-specific treatment effects from panel

Strategyproof Decision-Making in Panel Data Settings and Beyond

We consider the classical problem of decision-making using panel data , in which a decision-maker, called the principal, gets noisy, repeated measurements of multiple units (or agents). We consider a

Speed Up the Cold-Start Learning in Two-Sided Bandits with Many Arms

This work designs two-phase bandit algorithms that first use subsampling and low-rank matrix estimation to obtain a substantially smaller targeted set of products and then apply a UCB procedure on the target products to find the best one.

Counterfactual inference for sequential experimental design

We consider the problem of counterfactual inference in sequentially designed experiments wherein a collection of N units each undergo a sequence of interventions for T time periods, based on policies

Counterfactual inference for sequential experiments

A latent factors model is introduced over the counterfactual means that serves as a non-parametric generalization of the non-linear mixed effects model and the bilinear latent factor model considered in prior works to provide inference guarantees for thecounterfactual mean at the smallest possible scale.



Missing Not at Random in Matrix Completion: The Effectiveness of Estimating Missingness Probabilities Under a Low Nuclear Norm Assumption

This paper tackles MNAR matrix completion by solving a different matrix completion problem first that recovers missingness probabilities, and establishes finite-sample error bounds for how accurate these probability estimates are and how well these estimates debias standard matrix completion losses for the original matrix to be completed.

Modeling Dynamic Missingness of Implicit Feedback for Recommendation

This work proposes a latent variable named "user intent" to govern the temporal changes of item missingness, and a hidden Markov model to represent such a process, and explores two types of constraints to achieve a more compact and interpretable representation of user intents.

Estimation and Imputation in Probabilistic Principal Component Analysis with Missing Not At Random Data

An estimation of the loading coefficients and a data imputation method based on estimators of means, variances and covariances of missing variables, which prove identifiability of the PPCA parameters.

Matrix estimation by Universal Singular Value Thresholding

This paper introduces a simple estimation procedure, called Universal Singular Value Thresholding (USVT), that works for any matrix that has "a little bit of structure" and achieves the minimax error rate up to a constant factor.

The Power of Convex Relaxation: Near-Optimal Matrix Completion

This paper shows that, under certain incoherence assumptions on the singular vectors of the matrix, recovery is possible by solving a convenient convex program as soon as the number of entries is on the order of the information theoretic limit (up to logarithmic factors).

1-Bit Matrix Completion

A theory of matrix completion for the extreme case of noisy 1-bit observations is developed and it is shown that the maximum likelihood estimate under a suitable constraint returns an accurate estimate of M when ||M||_{\infty} <= \alpha, and rank(M) <= r.

Inference and uncertainty quantification for noisy matrix completion

A simple procedure to compensate for the bias of the widely used convex and nonconvex estimators and derive distributional characterizations for the resulting debiased estimators, which enable optimal construction of confidence intervals/regions for the missing entries and the low-rank factors.

Doubly Robust Joint Learning for Recommendation on Data Missing Not at Random

This work proposes an estimator that integrates the imputed errors and propensities in a doubly robust way to obtain unbiased performance estimation and alleviate the effect of the propensity variance.

Imputation and low-rank estimation with Missing Not At Random data

This paper proposes matrix completion methods to recover Missing Not At Random (MNAR) data to predict if the doctors should administrate tranexomic acid to patients with traumatic brain injury, that would limit excessive bleeding.

Matrix Completion With Data-Dependent Missingness Probabilities

Two new estimators are proposed, based on singular value thresholding and nuclear norm minimization, to recover the matrix under this assumption, and involve no tuning parameters, and are shown to be consistent under a low rank assumption.