• Corpus ID: 218870305

Causal Bayesian Optimization

  title={Causal Bayesian Optimization},
  author={Virginia Aglietti and Xiaoyu Lu and Andrei Paleyes and Javier Gonz' alez},
This paper studies the problem of globally optimizing a variable of interest that is part of a causal model in which a sequence of interventions can be performed. This problem arises in biology, operational research, communications and, more generally, in all fields where the goal is to optimize an output metric of a system of interconnected nodes. Our approach combines ideas from causal inference, uncertainty quantification and sequential decision making. In particular, it generalizes Bayesian… 
Dynamic Causal Bayesian Optimization
This paper gives theoretical results detailing how one can transfer interventional information across time steps and defines a dynamic causal GP model which can be used to quantify uncertainty and find optimal interventions in practice.
Estimating individual-level optimal causal interventions combining causal models and machine learning models
We introduce a new statistical causal inference method to estimate individual-level optimal causal intervention, that is, to which value we should set the value of a certain variable of an individual
Multi-task Causal Learning with Gaussian Processes
This paper proposes the first multi-task causal Gaussian process (GP) model, which it is called DAG-GP, that allows for information sharing across continuous interventions and across experiments on different variables and achieves the best fitting performance in a variety of real and synthetic settings.
BayesIMP: Uncertainty Quantification for Causal Data Fusion
Bayesian Interventional Mean Processes is introduced, a framework which combines ideas from probabilistic integration and kernel mean embeddings to represent interventional distributions in the reproducing kernel Hilbert space, while taking into account the uncertainty within each causal graph.
Chronological Causal Bandits
The Chronological Causal Bandit (CCB) is useful in discrete decision-making settings where the causal effects are changing across time and can be informed by earlier interventions in the same system.
BoGraph: Structured Bayesian Optimization From Logs for Systems with High-dimensional Parameter Space
BoGraph is proposed, a SBO framework that learns the system structure from its logs and transforms it into a probabilistic graph model, which allows the optimizer to find efficient configurations faster than other methods.
Supplementary material for Dynamic Causal Bayesian Optimisation
Results over multiple datasets and replicates are presented, showing no evidence of convergence between stationary and non-stationary DAG and SCM variables, and independent manipulative variables are identified.
BoGraph: structured bayesian optimization from logs for expensive systems with many parameters
BoGraph is an auto-tuning framework that builds a graph of the system components before optimizing it using causal structure learning and contextualizes the system via decomposition of the parameter space for faster convergence and handling of many parameters.
Automatic parameter selection for electron ptychography via Bayesian optimization
Electron ptychography provides new opportunities to resolve atomic structures with deep sub-angstrom spatial resolution and to study electron-beam sensitive materials with high dose efficiency. In
Multi-Objective Hyperparameter Optimization - An Overview
Overview FLORIAN KARL∗, Fraunhofer Institut für integrierte Schaltungen, Germany TOBIAS PIELOK∗, Ludwig-Maximilians-Universität München, Germany JULIA MOOSBAUER, Ludwig-Maximilians-Universität


Structural Causal Bandits: Where to Intervene?
This paper builds a new algorithm that takes as input a causal structure and finds a minimal, sound, and complete set of qualified arms that an agent should play to maximize its expected reward and empirically demonstrates that the new strategy learns an optimal policy and leads to orders of magnitude faster convergence rates when compared with its causal-insensitive counterparts.
Generalized Thompson sampling for sequential decision-making and causal inference
The results suggest that Thompson sampling might not merely be a useful heuristic, but a principled method to address problems of adaptive sequential decision-making and causal inference.
Structural Causal Bandits with Non-Manipulable Variables
A procedure is developed that takes as argument partially specified causal knowledge and identifies the possibly-optimal arms in structural bandits with non-manipulable variables and introduces an algorithm that uncovers non-trivial dependence structure among the Possibly-Optimal arms.
Causal Bandits: Learning Good Interventions via Causal Inference
A new algorithm is proposed that exploits the causal feedback and proves a bound on its simple regret that is strictly better (in all quantities) than algorithms that do not use the additional causal information.
Review of Causal Discovery Methods Based on Graphical Models
This paper aims to give a introduction to and a brief review of the computational methods for causal discovery that were developed in the past three decades, including constraint-based and score-based methods and those based on functional causal models, supplemented by some illustrations and applications.
Bandits with Unobserved Confounders: A Causal Approach
It is shown that to achieve low regret in certain realistic classes of bandit problems (namely, in the face of unobserved confounders), both experimental and observational quantities are required by the rational agent.
Causal Graph Analysis with the CAUSALGRAPH Procedure
The CAUSALGRAPH procedure, new in SAS/STAT® 15.1, for analyzing graphical causal models, is introduced and details about using directed acyclic graphs to represent and analyze a causal model are provided.
Woulda, Coulda, Shoulda: Counterfactually-Guided Policy Search
The Counterfactually-Guided Policy Search (CF-GPS) algorithm is proposed, which leverages structural causal models for counterfactual evaluation of arbitrary policies on individual off-policy episodes and can improve on vanilla model-based RL algorithms by making use of available logged data to de-bias model predictions.
Bayesian Optimization in a Billion Dimensions via Random Embeddings
Empirical results confirm that REMBO can effectively solve problems with billions of dimensions, provided the intrinsic dimensionality is low, and show thatREMBO achieves state-of-the-art performance in optimizing the 47 discrete parameters of a popular mixed integer linear programming solver.
Deconfounding Reinforcement Learning in Observational Settings
This work considers the problem of learning good policies solely from historical data in which unobserved factors affect both observed actions and rewards, and for the first time that confounders are taken into consideration for addressing full RL problems with observational data.