Search

Reinforcement learning provides a mathematical framework for learning-based control, whose success largely depends on the amount of data it can utilize. The efficient utilization of historical trajectories obtained from previous policies is essential for expediting policy optimization. Empirical evidence has shown that policy gradient methods based on importance sampling work well. However, existing literature often neglect the interdependence between trajectories from different iterations, and the good empirical performance lacks a rigorous theoretical justification. In this paper, we study a variant of the natural policy gradient method with reusing historical trajectories via importance sampling. We show that the bias of the proposed estimator of the gradient is asymptotically negligible, the resultant algorithm is convergent, and reusing past trajectories helps improve the convergence rate. We further apply the proposed estimator to popular policy optimization algorithms such as trust region policy optimization. Our theoretical results are verified on classical benchmarks.

Stochastic optimal control with unknown randomness distributions has been studied for a long time, encompassing robust control, distributionally robust control, and adaptive control. We propose a new episodic Bayesian approach that incorporates Bayesian learning with optimal control. In each episode, the approach learns the randomness distribution with a Bayesian posterior and subsequently solves the corresponding Bayesian average estimate of the true problem. The resulting policy is exercised during the episode, while additional data/observations of the randomness are collected to update the Bayesian posterior for the next episode. We show that the resulting episodic value functions and policies converge almost surely to their optimal counterparts of the true problem if the parametrized model of the randomness distribution is correctly specified. We further show that the asymptotic convergence rate of the episodic value functions is of the order $O(N^{-1/2})$. We develop an efficient computational method based on stochastic dual dynamic programming for a class of problems that have convex value functions. Our numerical results on a classical inventory control problem verify the theoretical convergence results and demonstrate the effectiveness of the proposed computational method.

We consider infinite-horizon Markov Decision Processes where parameters, such as transition probabilities, are unknown and estimated from data. The popular distributionally robust approach to addressing the parameter uncertainty can sometimes be overly conservative. In this paper, we utilize the recently proposed formulation, Bayesian risk Markov Decision Process (BR-MDP), to address parameter (or epistemic) uncertainty in MDPs (Lin et al., 2022). To solve the infinite-horizon BR-MDP with a class of convex risk measures, we propose a computationally efficient approach of approximate bilevel difference convex programming (ABDCP). The optimization is performed offline and produces the optimal policy that is represented as a finite state controller with desirable performance guarantees. We also demonstrate the empirical performance of the infinite-horizon BR-MDP formulation and proposed algorithms.

In this paper we consider the contextual multi-armed bandit problem for linear payoffs under a risk-averse criterion. At each round, contexts are revealed for each arm, and the decision maker chooses one arm to pull and receives the corresponding reward. In particular, we consider mean-variance as the risk criterion, and the best arm is the one with the largest mean-variance reward. We apply the Thompson Sampling algorithm for the disjoint model, and provide a comprehensive regret analysis for a variant of the proposed algorithm. For $T$ rounds, $K$ actions, and $d$-dimensional feature vectors, we prove a regret bound of $O((1+\rho+\frac{1}{\rho}) d\ln T \ln \frac{K}{\delta}\sqrt{d K T^{1+2\epsilon} \ln \frac{K}{\delta} \frac{1}{\epsilon}})$ that holds with probability $1-\delta$ under the mean-variance criterion with risk tolerance $\rho$, for any $0<\epsilon<\frac{1}{2}$, $0<\delta<1$. The empirical performance of our proposed algorithms is demonstrated via a portfolio selection problem.

Temperature drift, stress birefringence and low frequency vibration lead to the randomness and fluctuation of the output of optical voltage sensor(OVS). In order to solve the problem, this study adopts the lock-in amplifier technology with the aid of a high-speed rotating electrode to realize electric field modulation. This technology could shift the measured signal frequency band from near 50 Hz moved to several kilometer Hz, so as to make the output signal avoid the interference from low-frequency temperature drift, stress birefringence and vibration, leading to higher stability and reliability. The electro-optic coupling wave theory and static electric field finite element method are utilized to investigate the shape of modulation wave. The simulation results proves that lock-in technology is able to prevent the measured voltage signal from the large step signal interference and restore the perfect original signal. While the sample rate is decreased to the value of the modulation frequency., Comment: 6 pages, 11 figures, conference

We consider stochastic optimization under distributional uncertainty, where the unknown distributional parameter is estimated from streaming data that arrive sequentially over time. Moreover, data may depend on the decision of the time when they are generated. For both decision-independent and decision-dependent uncertainties, we propose an approach to jointly estimate the distributional parameter via Bayesian posterior distribution and update the decision by applying stochastic gradient descent on the Bayesian average of the objective function. Our approach converges asymptotically over time and achieves the convergence rates of classical SGD in the decision-independent case. We demonstrate the empirical performance of our approach on both synthetic test problems and a classical newsvendor problem.

To solve the problems of instability and thermal stress birefringence of voltage divided medium in optical voltage sensor (OVS), a half-wave voltage regulation method based on a central symmetry electrode and an arbitrary directional electric field is proposed in this paper. The method uses a center-symmetric copper foil electrode or the ITO (Indium Tin Oxide) transparent electrode to apply voltage, which can arbitrarily adjust the direction of the average electric field in the crystal, so as to improve the half-wave voltage of the bismuth germanate (Bi4Ge3O12, or BGO) crystal directly without additional SF6 gas or quartz glass. The internal electric field of the BGO crystal is simulated by the finite element method, and the half-wave voltage is calculated by the coupling wave theory of electro-optic effect. The results show that the maximum regulated half-wave voltage, considering the 1{\deg} angle, can reach 2.67e3 kV. And this is consistent to the experimental results 1.72e3 kV with difference of 35.6%. For the measurement of 110kV line voltage, that is, 63.5kV phase voltage, to ensure a 0.2 accuracy class, the angle error of incident light, and the polarization direction and the electric field direction is required to be less than 3.9', which can be achieved by the existing technologies.

We introduce a new framework, Bayesian Distributionally Robust Optimization (Bayesian-DRO), for data-driven stochastic optimization where the underlying distribution is unknown. Bayesian-DRO contrasts with most of the existing DRO approaches in the use of Bayesian estimation of the unknown distribution. To make computation of Bayesian updating tractable, Bayesian-DRO first assumes the underlying distribution takes a parametric form with unknown parameter and then computes the posterior distribution of the parameter. To address the model uncertainty brought by the assumed parametric distribution, Bayesian-DRO constructs an ambiguity set of distributions with the assumed parametric distribution as the reference distribution and then optimizes with respect to the worst case in the ambiguity set. We show the consistency of the Bayesian posterior distribution and subsequently the convergence of objective functions and optimal solutions of Bayesian-DRO. Our consistency result of the Bayesian posterior requires simpler assumptions than the classical literature on Bayesian consistency. We also consider several approaches for selecting the ambiguity set size in Bayesian-DRO and compare them numerically. Our numerical experiments demonstrate the out-of-sample performance of Bayesian-DRO in comparison with Kullback-Leibler-based (KL-) and Wasserstein-based empirical DRO as well as risk-neutral Bayesian Risk Optimization. Our numerical results shed light on how to choose the modeling framework (Bayesian-DRO, KL-DRO, Wasserstein-DRO) for specific problems, but the choice for general problems still remains an important and open question.

We consider finite-horizon Markov Decision Processes where parameters, such as transition probabilities, are unknown and estimated from data. The popular distributionally robust approach to addressing the parameter uncertainty can sometimes be overly conservative. In this paper, we propose a new formulation, Bayesian risk Markov Decision Process (BR-MDP), to address parameter uncertainty in MDPs, where a risk functional is applied in nested form to the expected total cost with respect to the Bayesian posterior distribution of the unknown parameters. The proposed formulation provides more flexible risk attitutes towards parameter uncertainty and takes into account the availability of data in future times stages. To solve the proposed formulation with the conditional value-at-risk (CVaR) risk functional, we propose an efficient approximation algorithm by deriving an analytical approximation of the value function and utilizing the convexity of CVaR. We demonstrate the empirical performance of the BR-MDP formulation and proposed algorithms on a gambler's betting problem and an inventory control problem.

Repeated asymptomatic screening for SARS-CoV-2 promises to control spread of the virus but would require too many resources to implement at scale. Group testing is promising for screening more people with fewer test resources: multiple samples tested together in one pool can be excluded with one negative test result. Existing approaches to group testing design for SARS-CoV-2 asymptomatic screening, however, do not consider dilution effects: that false negatives become more common with larger pools. As a consequence, they may recommend pool sizes that are too large or misestimate the benefits of screening. Modeling dilution effects, we derive closed-form expressions for the expected number of tests and false negative/positives per person screened under two popular group testing methods: the linear and square array methods. We find that test error correlation induced by a common viral load across an individual's samples results in many fewer false negatives than would be expected from less realistic but more widely assumed independent errors. This insight also suggests that false positives can be controlled through repeated tests without significantly increasing false negatives. Using these closed-form expressions to trace a Pareto frontier over error rates and tests, we design testing protocols for repeated asymptomatic screening of a large population. We minimize disease prevalence by optimizing a time-varying pool sizes and screening frequency constrained by daily test capacity and a false positive limit. This provides a testing protocol practitioners can use for mitigating COVID-19. In a case study, we demonstrate the effectiveness of this methodology in controlling spread.