# Combinatorial Causal Bandits

@article{Feng2022CombinatorialCB, title={Combinatorial Causal Bandits}, author={Shi Feng and W. Chen}, journal={ArXiv}, year={2022}, volume={abs/2206.01995} }

In combinatorial causal bandits (CCB), the learning agent chooses at most K variables in each round to intervene, col-lects feedback from the observed variables, with the goal of minimizing expected regret on the target variable Y . Different from all prior studies on causal bandits, CCB needs to deal with exponentially large action space. We study under the context of binary generalized linear models (BGLMs) with a succinct parametric representation of the causal models. We present the…

## 2 Citations

### Pure Exploration of Causal Bandits

- Computer ScienceArXiv
- 2022

This work provides first gap-dependent fully adaptive fully adaptive pure exploration algorithms on three types of causal models including parallel graphs, general graphs with small number of backdoor parents, and binary generalized linear models.

### Causal Bandits for Linear Structural Equation Models

- Computer ScienceArXiv
- 2022

This paper studies the problem of designing an optimal sequence of interventions in a causal graphical model to minimize the cumulative regret with respect to the best intervention in hindsight and proposes two algorithms for the frequentist (UCB-based) and Bayesian settings.

## References

SHOWING 1-10 OF 38 REFERENCES

### Causality

- Philosophy
- 1938

IT is a rather depressing task for a Catholic to write a technical philosophical essay, for it is improbable that the philosophers whom he criticises wi l l read it . F r Hawkins therefore deserves…

### Matrix anti-concentration inequalities with applications

- Computer Science, MathematicsSTOC
- 2022

Two matrix anti-concentration inequalities are established, which lower bound the minimum singular values of the sum of independent positive semidefinite self-adjoint matrices and the linear combination of independent random matrices with independent Gaussian coefficients.

### Online Influence Maximization with Node-level Feedback Using Standard Offline Oracles

- Computer ScienceAAAI
- 2022

A novel adaptation of the maximum likelihood estimation (MLE) approach is applied to learn the graph parameters and its confidence region (a confidence ellipsoid) to resolve the challenges for the famous independent cascade (IC) diffusion model.

### Online Influence Maximization under Linear Threshold Model

- Computer ScienceNeurIPS
- 2020

This paper addresses OIM in the linear threshold (LT) model, incorporating ideas from linear bandits and design an algorithm LT-LinUCB that is consistent with the observed feedback and provides an algorithm OIM-ETC with regret bound, which is model-independent, simple and has less requirement on online feedback and offline computation.

### Provably Optimal Algorithms for Generalized Linear Contextual Bandits

- Computer ScienceICML
- 2017

This work proposes an upper confidence bound based algorithm for generalized linear contextual bandits, which achieves an \tilde{O}(\sqrt{dT}) regret over T rounds with d dimensional feature vectors, and proves it to have optimal regret for the certain cases.

### Combinatorial Multi-Armed Bandit and Its Extension to Probabilistically Triggered Arms

- Computer ScienceJ. Mach. Learn. Res.
- 2016

The regret analysis is tight in that it matches the bound of UCB1 algorithm (up to a constant factor) for the classical MAB problem, and it significantly improves the regret bound in an earlier paper on combinatorial bandits with linear rewards.

### Parametric Bandits: The Generalized Linear Case

- Computer ScienceNIPS
- 2010

The analysis highlights a key difficulty in generalizing linear bandit algorithms to the non-linear case, which is solved in GLM-UCB by focusing on the reward space rather than on the parameter space, and provides a tuning method based on asymptotic arguments, which leads to significantly better practical performance.

### Regret Analysis of Bandit Problems with Causal Background Knowledge

- Computer ScienceUAI
- 2020

It is observed that even with a few hundreds of iterations, the regret of causal algorithms is less than that of standard algorithms by a factor of three, and under certain causal structures, these algorithms scale better than the standard bandit algorithms as the number of interventions increases.

### Budgeted and nonbudgeted causal bandits

- International Conference on Artificial Intelligence and Statistics, 2017–2025. PMLR.
- 2021

### Artificial Intelligence A Modern Approach Global Edition

- Computer Science
- 2022

Artificial Intelligence: A Modern Approach, eBook, Global Edition, and the Hundred-page Machine Learning Book are available.