اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدCombining Offline Causal Inference and Online Bandit Learning for Data Driven Decision
298
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
A fundamental question for companies with large amount of logged data is: How to use such logged data together with incoming streaming data to make good decisions? Many companies currently make decisions via online A/B tests, but wrong decisions during testing hurt users experiences and cause irreversible damage. A typical alternative is offline causal inference, which analyzes logged data alone to make decisions. However, these decisions are not adaptive to the new incoming data, and so a wrong decision will continuously hurt users experiences. To overcome the aforementioned limitations, we propose a framework to unify offline causal inference algorithms (e.g., weighting, matching) and online learning algorithms (e.g., UCB, LinUCB). We propose novel algorithms and derive bounds on the decision accuracy via the notion of regret. We derive the first upper regret bound for forest-based online bandit algorithms. Experiments on two real datasets show that our algorithms outperform other algorithms that use only logged data or online feedbacks, or algorithms that do not use the data properly.
قيم البحث
اقرأ أيضاً
We study a novel variant of online finite-horizon Markov Decision Processes with adversarially changing loss functions and initially unknown dynamics. In each episode, the learner suffers the loss accumulated along the trajectory realized by the poli
cy chosen for the episode, and observes aggregate bandit feedback: the trajectory is revealed along with the cumulative loss suffered, rather than the individual losses encountered along the trajectory. Our main result is a computationally efficient algorithm with $O(sqrt{K})$ regret for this setting, where $K$ is the number of episodes. We establish this result via an efficient reduction to a novel bandit learning setting we call Distorted Linear Bandits (DLB), which is a variant of bandit linear optimization where actions chosen by the learner are adversarially distorted before they are committed. We then develop a computationally-efficient online algorithm for DLB for which we prove an $O(sqrt{T})$ regret bound, where $T$ is the number of time steps. Our algorithm is based on online mirror descent with a self-concordant barrier regularization that employs a novel increasing learning rate schedule.
This paper deals with bandit online learning problems involving feedback of unknown delay that can emerge in multi-armed bandit (MAB) and bandit convex optimization (BCO) settings. MAB and BCO require only values of the objective function involved th
at become available through feedback, and are used to estimate the gradient appearing in the corresponding iterative algorithms. Since the challenging case of feedback with emph{unknown} delays prevents one from constructing the sought gradient estimates, existing MAB and BCO algorithms become intractable. For such challenging setups, delayed exploration, exploitation, and exponential (DEXP3) iterations, along with delayed bandit gradient descent (DBGD) iterations are developed for MAB and BCO, respectively. Leveraging a unified analysis framework, it is established that the regret of DEXP3 and DBGD are ${cal O}big( sqrt{Kbar{d}(T+D)} big)$ and ${cal O}big( sqrt{K(T+D)} big)$, respectively, where $bar{d}$ is the maximum delay and $D$ denotes the delay accumulated over $T$ slots. Numerical tests using both synthetic and real data validate the performance of DEXP3 and DBGD.
We formulate a new problem at the intersectionof semi-supervised learning and contextual bandits,motivated by several applications including clini-cal trials and ad recommendations. We demonstratehow Graph Convolutional Network (GCN), a semi-supervis
ed learning approach, can be adjusted tothe new problem formulation. We also propose avariant of the linear contextual bandit with semi-supervised missing rewards imputation. We thentake the best of both approaches to develop multi-GCN embedded contextual bandit. Our algorithmsare verified on several real world datasets.
Recent theoretical work studies sample-efficient reinforcement learning (RL) extensively in two settings: learning interactively in the environment (online RL), or learning from an offline dataset (offline RL). However, existing algorithms and theori
es for learning near-optimal policies in these two settings are rather different and disconnected. Towards bridging this gap, this paper initiates the theoretical study of policy finetuning, that is, online RL where the learner has additional access to a reference policy $mu$ close to the optimal policy $pi_star$ in a certain sense. We consider the policy finetuning problem in episodic Markov Decision Processes (MDPs) with $S$ states, $A$ actions, and horizon length $H$. We first design a sharp offline reduction algorithm -- which simply executes $mu$ and runs offline policy optimization on the collected dataset -- that finds an $varepsilon$ near-optimal policy within $widetilde{O}(H^3SC^star/varepsilon^2)$ episodes, where $C^star$ is the single-policy concentrability coefficient between $mu$ and $pi_star$. This offline result is the first that matches the sample complexity lower bound in this setting, and resolves a recent open question in offline RL. We then establish an $Omega(H^3Smin{C^star, A}/varepsilon^2)$ sample complexity lower bound for any policy finetuning algorithm, including those that can adaptively explore the environment. This implies that -- perhaps surprisingly -- the optimal policy finetuning algorithm is either offline reduction or a purely online RL algorithm that does not use $mu$. Finally, we design a new hybrid offline/online algorithm for policy finetuning that achieves better sample complexity than both vanilla offline reduction and purely online RL algorithms, in a relaxed setting where $mu$ only satisfies concentrability partially up to a certain time step.
In this paper, we consider the contextual variant of the MNL-Bandit problem. More specifically, we consider a dynamic set optimization problem, where in every round a decision maker offers a subset (assortment) of products to a consumer, and observes
their response. Consumers purchase products so as to maximize their utility. We assume that the products are described by a set of attributes and the mean utility of a product is linear in the values of these attributes. We model consumer choice behavior by means of the widely used Multinomial Logit (MNL) model, and consider the decision makers problem of dynamically learning the model parameters, while optimizing cumulative revenue over the selling horizon $T$. Though this problem has attracted considerable attention in recent times, many existing methods often involve solving an intractable non-convex optimization problem and their theoretical performance guarantees depend on a problem dependent parameter which could be prohibitively large. In particular, existing algorithms for this problem have regret bounded by $O(sqrt{kappa d T})$, where $kappa$ is a problem dependent constant that can have exponential dependency on the number of attributes. In this paper, we propose an optimistic algorithm and show that the regret is bounded by $O(sqrt{dT} + kappa)$, significantly improving the performance over existing methods. Further, we propose a convex relaxation of the optimization step which allows for tractable decision-making while retaining the favourable regret guarantee.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
