Tractable contextual bandits beyond realizability

by   Sanath Kumar Krishnamurthy, et al.

Tractable contextual bandit algorithms often rely on the realizability assumption – i.e., that the true expected reward model belongs to a known class, such as linear functions. We investigate issues that arise in the absence of realizability and note that the dynamics of adaptive data collection can lead commonly used bandit algorithms to learn a suboptimal policy. In this work, we present a tractable bandit algorithm that is not sensitive to the realizability assumption and computationally reduces to solving a constrained regression problem in every epoch. When realizability does not hold, our algorithm ensures the same guarantees on regret achieved by realizability-based algorithms under realizability, up to an additive term that accounts for the misspecification error. This extra term is proportional to T times the (2/5)-root of the mean squared error between the best model in the class and the true model, where T is the total number of time-steps. Our work sheds light on the bias-variance trade-off for tractable contextual bandits. This trade-off is not captured by algorithms that assume realizability, since under this assumption there exists an estimator in the class that attains zero bias.


page 1

page 2

page 3

page 4


Model Selection for Generic Contextual Bandits

We consider the problem of model selection for the general stochastic co...

Adapting to misspecification in contextual bandits with offline regression oracles

Computationally efficient contextual bandits are often based on estimati...

Universal and data-adaptive algorithms for model selection in linear contextual bandits

Model selection in contextual bandits is an important complementary prob...

Bypassing the Monster: A Faster and Simpler Optimal Algorithm for Contextual Bandits under Realizability

We consider the general (stochastic) contextual bandit problem under the...

Efficient and Robust Algorithms for Adversarial Linear Contextual Bandits

We consider an adversarial variant of the classic K-armed linear context...

Mitigating Bias in Adaptive Data Gathering via Differential Privacy

Data that is gathered adaptively --- via bandit algorithms, for example ...

A bandit-learning approach to multifidelity approximation

Multifidelity approximation is an important technique in scientific comp...