Estimating value functions is a core component of reinforcement learning algorithms. Temporal difference (TD) learning algorithms use bootstrapping, i.e. they update the value function toward a learning target using value estimates at subsequent time-steps. Alternatively, the value function can be updated toward a learning target constructed by separately predicting successor features (SF)--a policy-dependent model--and linearly combining them with instantaneous rewards. We focus on bootstrapping targets used when estimating value functions, and propose a new backup target, the $η$-return mixture, which implicitly combines value-predictive knowledge (used by TD methods) with (successor) feature-predictive knowledge--with a parameter $η$ capturing how much to rely on each. We illustrate that incorporating predictive knowledge through an $ηγ$-discounted SF model makes more efficient use of sampled experience, compared to either extreme, i.e. bootstrapping entirely on the value function estimate, or bootstrapping on the product of separately estimated successor features and instantaneous reward models. We empirically show this approach leads to faster policy evaluation and better control performance, for tabular and nonlinear function approximations, indicating scalability and generality.
This paper considers Safe Policy Improvement (SPI) in Batch Reinforcement Learning (Batch RL): from a fixed dataset and without direct access to the true environment, train a policy that is guaranteed to perform at least as well as the baseline policy used to collect the data. Our approach, called SPI with Baseline Bootstrapping (SPIBB), is inspired by the knows-what-it-knows paradigm: it bootstraps the trained policy with the baseline when the uncertainty is high. Our first algorithm, $Π_b$-SPIBB, comes with SPI theoretical guarantees. We also implement a variant, $Π_{\leq b}$-SPIBB, that is …
Tsinghua University（清华大学）；Stanford University（斯坦福大学）；University of Washington（华盛顿大学）
This paper presents a new model-free algorithm for episodic finite-horizon Markov Decision Processes (MDP), Adaptive Multi-step Bootstrap (AMB), which enjoys a stronger gap-dependent regret bound. The first innovation is to estimate the optimal $Q$-function by combining an optimistic bootstrap with an adaptive multi-step Monte Carlo rollout. The second innovation is to select the action with the largest confidence interval length among admissible actions that are not dominated by any other actions. We show when each state has a unique optimal action, AMB achieves a gap-dependent regret bound t…
Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Tencent（腾讯）
The exploration/exploitation (E&E) dilemma lies at the core of interactive systems such as online advertising, for which contextual bandit algorithms have been proposed. Bayesian approaches provide guided exploration with principled uncertainty estimation, but the applicability is often limited due to over-simplified assumptions. Non-Bayesian bootstrap methods, on the other hand, can apply to complex problems by using deep reward models, but lacks clear guidance to the exploration behavior. It still remains largely unsolved to develop a practical method for complex deep contextual bandits. In …
US Army Research Laboratory（美国陆军研究实验室）；Texas A&M University（得克萨斯农工大学）；Columbia University（哥伦比亚大学）
This paper investigates how to utilize different forms of human interaction to safely train autonomous systems in real-time by learning from both human demonstrations and interventions. We implement two components of the Cycle-of-Learning for Autonomous Systems, which is our framework for combining multiple modalities of human interaction. The current effort employs human demonstrations to teach a desired behavior via imitation learning, then leverages intervention data to correct for undesired behaviors produced by the imitation learner to teach novel tasks to an autonomous agent safely, afte…
BFMT: Enhancing Search Capabilities of Tree Sampler via Bootstrap Flow-Map Tree
Department of CSE, Washington University in St.Louis, USA（美国圣路易斯华盛顿大学计算机科学与工程系）；Department of Information Technology, Uppsala University, Sweden（瑞典乌普萨拉大学信息技术系）
In machine learning, an agent needs to estimate uncertainty to efficiently explore and adapt and to make effective decisions. A common approach to uncertainty estimation maintains an ensemble of models. In recent years, several approaches have been proposed for training ensembles, and conflicting views prevail with regards to the importance of various ingredients of these approaches. In this paper, we aim to address the benefits of two ingredients -- prior functions and bootstrapping -- which have come into question. We show that prior functions can significantly improve an ensemble agent's jo…
We introduce a new unsupervised pretraining objective for reinforcement learning. During the unsupervised reward-free pretraining phase, the agent maximizes mutual information between tasks and states induced by the policy. Our key contribution is a novel lower bound of this intractable quantity. We show that by reinterpreting and combining variational successor features~\citep{Hansen2020Fast} with nonparametric entropy maximization~\citep{liu2021behavior}, the intractable mutual information can be efficiently optimized. The proposed method Active Pretraining with Successor Feature (APS) explo…
Centrum voor Wiskunde en Informatica (CWI)（荷兰国家数学与计算机科学研究院（CWI））
We show how models for prediction with expert advice can be defined concisely and clearly using hidden Markov models (HMMs); standard HMM algorithms can then be used to efficiently calculate, among other things, how the expert predictions should be weighted according to the model. We cast many existing models as HMMs and recover the best known running times in each case. We also describe two new models: the switch distribution, which was recently developed to improve Bayesian/Minimum Description Length model selection, and a new generalisation of the fixed share algorithm based on run-length c…
University of Southern California（南加州大学）；Boston University（波士顿大学）；Massachusetts Institute of Technology（麻省理工学院）；The NSF Institute for Artificial Intelligence and Fundamental Interactions（美国国家科学基金会人工智能与基本相互作用研究所）
In this paper, we propose a novel weighted combination feature selection method using bootstrap and fuzzy sets. The proposed method mainly consists of three processes, including fuzzy sets generation using bootstrap, weighted combination of fuzzy sets and feature ranking based on defuzzification. We implemented the proposed method by combining four state-of-the-art feature selection methods and evaluated the performance based on three publicly available biomedical datasets using five-fold cross validation. Based on the feature selection results, our proposed method produced comparable (if not …
University of Trento（特伦托大学）；Tomsk Polytechnic University（托木斯克理工大学）；UNSW（新南威尔士大学）
This paper discusses how crowd and machine classifiers can be efficiently combined to screen items that satisfy a set of predicates. We show that this is a recurring problem in many domains, present machine-human (hybrid) algorithms that screen items efficiently and estimate the gain over human-only or machine-only screening in terms of performance and cost. We further show how, given a new classification problem and a set of classifiers of unknown accuracy for the problem at hand, we can identify how to manage the cost-accuracy trade off by progressively determining if we should spend budget …
