Holdout validation and hyperparameter tuning from data is a long-standing problem in offline reinforcement learning (RL). A standard framework is to use off-policy evaluation (OPE) methods to evaluate and select the policies, but OPE either incurs exponential variance (e.g., importance sampling) or has hyperparameters on their own (e.g., FQE and model-based). We focus on hyperparameter tuning for OPE itself, which is even more under-investigated. Concretely, we select among candidate value functions ("model-free") or dynamics ("model-based") to best assess the performance of a target policy. Concretely, we select among candidate value functions (``model-free'') or dynamics models (``model-based'') to best assess the performance of a target policy. We develop: (1) new model-free and model-based selectors with theoretical guarantees, and (2) a new experimental protocol for empirically evaluating them. Compared to the model-free protocol in prior works, our new protocol allows for more stable generation and better control of candidate value functions in an optimization-free manner, and evaluation of model-free and model-based methods alike. We exemplify the protocol on Gym-Hopper, and find that our new model-free selector, LSTD-Tournament, demonstrates promising empirical performance.
This paper addresses the problem of policy selection in domains with abundant logged data, but with a restricted interaction budget. Solving this problem would enable safe evaluation and deployment of offline reinforcement learning policies in industry, robotics, and recommendation domains among others. Several off-policy evaluation (OPE) techniques have been proposed to assess the value of policies using only logged data. However, there is still a big gap between the evaluation by OPE and the full online evaluation. Yet, large amounts of online interactions are often not possible in practice.…
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
How to select between policies and value functions produced by different training algorithms in offline reinforcement learning (RL) -- which is crucial for hyperpa-rameter tuning -- is an important open question. Existing approaches based on off-policy evaluation (OPE) often require additional function approximation and hence hyperparameters, creating a chicken-and-egg situation. In this paper, we design hyperparameter-free algorithms for policy selection based on BVFT [XJ21], a recent theoretical advance in value-function selection, and demonstrate their effectiveness in discrete-action bench…
We study the problem of model selection in batch policy optimization: given a fixed, partial-feedback dataset and $M$ model classes, learn a policy with performance that is competitive with the policy derived from the best model class. We formalize the problem in the contextual bandit setting with linear model classes by identifying three sources of error that any model selection algorithm should optimally trade-off in order to be competitive: (1) approximation error, (2) statistical complexity, and (3) coverage. The first two sources are common in model selection for supervised learning, wher…
We consider off-policy evaluation in the contextual bandit setting for the purpose of obtaining a robust off-policy selection strategy, where the selection strategy is evaluated based on the value of the chosen policy in a set of proposal (target) policies. We propose a new method to compute a lower bound on the value of an arbitrary target policy given some logged data in contextual bandits for a desired coverage. The lower bound is built around the so-called Self-normalized Importance Weighting (SN) estimator. It combines the use of a semi-empirical Efron-Stein tail inequality to control the…
We develop a generic data-driven method for estimator selection in off-policy policy evaluation settings. We establish a strong performance guarantee for the method, showing that it is competitive with the oracle estimator, up to a constant factor. Via in-depth case studies in contextual bandits and reinforcement learning, we demonstrate the generality and applicability of the method. We also perform comprehensive experiments, demonstrating the empirical efficacy of our approach and comparing with related approaches. In both case studies, our method compares favorably with existing methods.
One of the questions that arises when designing models that learn to solve multiple tasks simultaneously is how much of the available training budget should be devoted to each individual task. We refer to any formalized approach to addressing this problem (learned or otherwise) as a task selection policy. In this work we provide an empirical evaluation of the performance of some common task selection policies in a synthetic bandit-style setting, as well as on the GLUE benchmark for natural language understanding. We connect task selection policy learning to existing work on automated curriculu…
University of Alabama in Huntsville（阿拉巴马大学亨茨维尔分校）；Columbia University（哥伦比亚大学）
Machine learning models have widely been used in fraud detection systems. Most of the research and development efforts have been concentrated on improving the performance of the fraud scoring models. Yet, the downstream fraud alert systems still have limited to no model adoption and rely on manual steps. Alert systems are pervasively used across all payment channels in retail banking and play an important role in the overall fraud detection process. Current fraud detection systems end up with large numbers of dropped alerts due to their inability to account for the alert processing capacity. I…
Evolutionary Algorithms (EAs) and Deep Reinforcement Learning (DRL) have recently been integrated to take the advantage of the both methods for better exploration and exploitation.The evolutionary part in these hybrid methods maintains a population of policy networks.However, existing methods focus on optimizing the parameters of policy network, which is usually high-dimensional and tricky for EA.In this paper, we shift the target of evolution from high-dimensional parameter space to low-dimensional action space.We propose Evolutionary Action Selection-Twin Delayed Deep Deterministic Policy Gr…
Genetic algorithms have been widely used in many practical optimization problems. Inspired by natural selection, operators, including mutation, crossover and selection, provide effective heuristics for search and black-box optimization. However, they have not been shown useful for deep reinforcement learning, possibly due to the catastrophic consequence of parameter crossovers of neural networks. Here, we present Genetic Policy Optimization (GPO), a new genetic algorithm for sample-efficient deep policy optimization. GPO uses imitation learning for policy crossover in the state space and appli…
Warwick Manufacturing Group (WMG), University of Warwick（华威大学华威制造集团（WMG））；University of Warwick（华威大学）；United Kingdom（英国（注意：英国是国家，非机构，这里可能是原文本中地址信息，按要求不提取））
Policy gradient reinforcement learning (RL) algorithms have achieved impressive performance in challenging learning tasks such as continuous control, but suffer from high sample complexity. Experience replay is a commonly used approach to improve sample efficiency, but gradient estimators using past trajectories typically have high variance. Existing sampling strategies for experience replay like uniform sampling or prioritised experience replay do not explicitly try to control the variance of the gradient estimates. In this paper, we propose an online learning algorithm, adaptive experience s…
Learning to take actions based on observations is a core requirement for artificial agents to be able to be successful and robust at their task. Reinforcement Learning (RL) is a well-known technique for learning such policies. However, current RL algorithms often have to deal with reward shaping, have difficulties generalizing to other environments and are most often sample inefficient. In this paper, we explore active inference and the free energy principle, a normative theory from neuroscience that explains how self-organizing biological systems operate by maintaining a model of the world an…
Equilibrium Selection in Multi-Agent Policy Gradients via Opponent-Aware Basin Entry
Bloomsbury Technology（布洛姆斯伯里技术）；London School of Economics and Political Science（伦敦政治经济学院）；University of Bristol（布里斯托大学）；Johannes Kepler University Linz（林茨约翰尼斯·开普勒大学）；Odesa Polytechnic National University（敖德萨国立技术大学）
Monte Carlo Tree Search (MCTS) is a sampling best-first method to search for optimal decisions. The success of MCTS depends heavily on how the MCTS statistical tree is built and the selection policy plays a fundamental role in this. A particular selection policy that works particularly well, widely adopted in MCTS, is the Upper Confidence Bounds for Trees, referred to as UCT. Other more sophisticated bounds have been proposed by the community with the goal to improve MCTS performance on particular problems. Thus, it is evident that while the MCTS UCT behaves generally well, some variants might…
Sri Lanka Technological Campus（斯里兰卡科技园区）；University of Peradeniya（佩拉德尼亚大学）
This paper considers a multi-armed bandit (MAB) problem in which multiple mobile agents receive rewards by sampling from a collection of spatially dispersed stochastic processes, called bandits. The goal is to formulate a decentralized policy for each agent, in order to maximize the total cumulative reward over all agents, subject to option availability and inter-agent communication constraints. The problem formulation is motivated by applications in which a team of autonomous mobile robots cooperates to accomplish an exploration and exploitation task in an uncertain environment. Bandit locati…
Inria（法国国家信息与自动化研究所）；CNRS（法国国家科学研究中心）；Université de Lorraine（洛林大学）
One of the most interesting features of Bayesian optimization for direct policy search is that it can leverage priors (e.g., from simulation or from previous tasks) to accelerate learning on a robot. In this paper, we are interested in situations for which several priors exist but we do not know in advance which one fits best the current situation. We tackle this problem by introducing a novel acquisition function, called Most Likely Expected Improvement (MLEI), that combines the likelihood of the priors and the expected improvement. We evaluate this new acquisition function on a transfer lear…
DeepMind（深度思考）；Princeton University（普林斯顿大学）；University of Alberta（阿尔伯塔大学）
This paper provides a statistical analysis of high-dimensional batch Reinforcement Learning (RL) using sparse linear function approximation. When there is a large number of candidate features, our result sheds light on the fact that sparsity-aware methods can make batch RL more sample efficient. We first consider the off-policy policy evaluation problem. To evaluate a new target policy, we analyze a Lasso fitted Q-evaluation method and establish a finite-sample error bound that has no polynomial dependence on the ambient dimension. To reduce the Lasso bias, we further propose a post model-sele…
University of Catania（卡塔尼亚大学）；University of Oulu（奥卢大学）
To tackle the heterogeneous requirements of beyond 5G (B5G) and future 6G wireless networks, conventional medium access control (MAC) procedures need to evolve to enable base stations (BSs) and user equipments (UEs) to automatically learn innovative MAC protocols catering to extremely diverse services. This topic has received significant attention, and several reinforcement learning (RL) algorithms, in which BSs and UEs are cast as agents, are available with the aim of learning a communication policy based on agents' local observations. However, current approaches are typically overfitted to t…
Massachusetts Institute of Technology（麻省理工学院）；Microsoft Research NYC（微软研究院纽约分院）；University of Southern California（南加州大学）
We introduce the problem of model selection for contextual bandits, where a learner must adapt to the complexity of the optimal policy while balancing exploration and exploitation. Our main result is a new model selection guarantee for linear contextual bandits. We work in the stochastic realizable setting with a sequence of nested linear policy classes of dimension $d_1 < d_2 < \ldots$, where the $m^\star$-th class contains the optimal policy, and we design an algorithm that achieves $\tilde{O}(T^{2/3}d^{1/3}_{m^\star})$ regret with no prior knowledge of the optimal dimension $d_{m^\star}$. T…
Georgia Institute of Technology（佐治亚理工学院）；Argonne National Laboratory（阿贡国家实验室）
With the increasing penetration of distributed energy resources, distributed optimization algorithms have attracted significant attention for power systems applications due to their potential for superior scalability, privacy, and robustness to a single point-of-failure. The Alternating Direction Method of Multipliers (ADMM) is a popular distributed optimization algorithm; however, its convergence performance is highly dependent on the selection of penalty parameters, which are usually chosen heuristically. In this work, we use reinforcement learning (RL) to develop an adaptive penalty paramet…
Recent progress in model selection raises the question of the fundamental limits of these techniques. Under specific scrutiny has been model selection for general contextual bandits with nested policy classes, resulting in a COLT2020 open problem. It asks whether it is possible to obtain simultaneously the optimal single algorithm guarantees over all policies in a nested sequence of policy classes, or if otherwise this is possible for a trade-off $α\in[\frac{1}{2},1)$ between complexity term and time: $\ln(|Π_m|)^{1-α}T^α$. We give a disappointing answer to this question. Even in the purely st…
Learning personalized decision policies that generalize to the target population is of great relevance. Since training data is often not representative of the target population, standard policy learning methods may yield policies that do not generalize target population. To address this challenge, we propose a novel framework for learning policies that generalize to the target population. For this, we characterize the difference between the training data and the target population as a sample selection bias using a selection variable. Over an uncertainty set around this selection variable, we o…
University of Alberta（阿尔伯塔大学）；Google Research Brain Team（谷歌研究大脑团队）
The presence of uncertainty in policy evaluation significantly complicates the process of policy ranking and selection in real-world settings. We formally consider offline policy selection as learning preferences over a set of policy prospects given a fixed experience dataset. While one can select or rank policies based on point estimates of their policy values or high-confidence intervals, access to the full distribution over one's belief of the policy value enables more flexible selection algorithms under a wider range of downstream evaluation metrics. We propose BayesDICE for estimating thi…
Off-policy evaluation (OPE) aims to accurately evaluate the performance of counterfactual policies using only offline logged data. Although many estimators have been developed, there is no single estimator that dominates the others, because the estimators' accuracy can vary greatly depending on a given OPE task such as the evaluation policy, number of actions, and noise level. Thus, the data-driven estimator selection problem is becoming increasingly important and can have a significant impact on the accuracy of OPE. However, identifying the most accurate estimator using only the logged data i…
VERSES Research Labs（VERSES研究实验室）；Monash University（莫纳什大学）；Université du Québec à Montréal（魁北克大学蒙特利尔分校）
We develop an approach to policy selection in active inference that allows us to efficiently search large policy spaces by mapping each policy to its embedding in a vector space. We sample the expected free energy of representative points in the space, then perform a more thorough policy search around the most promising point in this initial sample. We consider various approaches to creating the policy embedding space, and propose using k-means clustering to select representative points. We apply our technique to a goal-oriented graph-traversal problem, for which naive policy selection is intr…
Selection of the Best Policy under Fairness Constraints for Subpopulations
Department of Industrial Engineering and Operations Research（工业工程与运营管理系）；Berkeley Artificial Intelligence Research Lab（伯克利人工智能研究实验室）；University of California Berkeley（加州大学伯克利分校）
Indian Institute of Technology Madras（印度理工学院马德拉斯分校）；Chennai Mathematical Institute（金奈数学研究所）；Indian Institute of Technology (ISM) Dhanbad（印度理工学院（印度矿业学院）丹巴德分校）；Tata Institute of Fundamental Research（塔塔基础研究所）
We revisit the classic problem of optimal subset selection in the online learning set-up. Assume that the set $[N]$ consists of $N$ distinct elements. On the $t$th round, an adversary chooses a monotone reward function $f_t: 2^{[N]} \to \mathbb{R}_+$ that assigns a non-negative reward to each subset of $[N].$ An online policy selects (perhaps randomly) a subset $S_t \subseteq [N]$ consisting of $k$ elements before the reward function $f_t$ for the $t$th round is revealed to the learner. As a consequence of its choice, the policy receives a reward of $f_t(S_t)$ on the $t$th round. Our goal is t…
University of Massachusetts（马萨诸塞大学）；University of Wisconsin（威斯康星大学）
We present a novel $l_1$ regularized off-policy convergent TD-learning method (termed RO-TD), which is able to learn sparse representations of value functions with low computational complexity. The algorithmic framework underlying RO-TD integrates two key ideas: off-policy convergent gradient TD methods, such as TDC, and a convex-concave saddle-point formulation of non-smooth convex optimization, which enables first-order solvers and feature selection using online convex regularization. A detailed theoretical and experimental analysis of RO-TD is presented. A variety of experiments are present…
University of Washington（华盛顿大学）；Monash University（莫纳什大学）；Ghent University（根特大学）
Existing work on privacy-preserving machine learning with Secure Multiparty Computation (MPC) is almost exclusively focused on model training and on inference with trained models, thereby overlooking the important data pre-processing stage. In this work, we propose the first MPC based protocol for private feature selection based on the filter method, which is independent of model training, and can be used in combination with any MPC protocol to rank features. We propose an efficient feature scoring protocol based on Gini impurity to this end. To demonstrate the feasibility of our approach for …
