arXiv subjects

Ming Yin

Publications and source records attributed to Ming Yin.

At least 91 records · Page 5Linked to original sources

arXiv2023

Non-stationary Reinforcement Learning under General Function Approximation

General function approximation is a powerful tool to handle large state and action spaces in a broad range of reinforcement learning (RL) scenarios. However, theoretical understanding of non-stationary MDPs with general function approximation is still limited. In this paper, we make the first such an attempt. We first propose a new complexity metric called dynamic Bellman Eluder (DBE) dimension for non-stationary MDPs, which subsumes majority of existing tractable RL problems in static MDPs as well as non-stationary MDPs. Based on the proposed complexity metric, we propose a novel confidence-set based model-free algorithm called SW-OPEA, which features a sliding window mechanism and a new confidence set design for non-stationary MDPs. We then establish an upper bound on the dynamic regret for the proposed algorithm, and show that SW-OPEA is provably efficient as long as the variation budget is not significantly large. We further demonstrate via examples of non-stationary linear and tabular MDPs that our algorithm performs better in small variation budget scenario than the existing UCB-type algorithms. To the best of our knowledge, this is the first dynamic regret analysis in non-stationary MDPs with general function approximation.

cs.LG↗

arXiv2023

Logarithmic Switching Cost in Reinforcement Learning beyond Linear MDPs

In many real-life reinforcement learning (RL) problems, deploying new policies is costly. In those scenarios, algorithms must solve exploration (which requires adaptivity) while switching the deployed policy sparsely (which limits adaptivity). In this paper, we go beyond the existing state-of-the-art on this problem that focused on linear Markov Decision Processes (MDPs) by considering linear Bellman-complete MDPs with low inherent Bellman error. We propose the ELEANOR-LowSwitching algorithm that achieves the near-optimal regret with a switching cost logarithmic in the number of episodes and linear in the time-horizon $H$ and feature dimension $d$. We also prove a lower bound proportional to $dH$ among all algorithms with sublinear regret. In addition, we show the ``doubling trick'' used in ELEANOR-LowSwitching can be further leveraged for the generalized linear function approximation, under which we design a sample-efficient algorithm with near-optimal switching cost.

cs.LG↗

arXiv2023

On Instance-Dependent Bounds for Offline Reinforcement Learning with Linear Function Approximation

Sample-efficient offline reinforcement learning (RL) with linear function approximation has recently been studied extensively. Much of prior work has yielded the minimax-optimal bound of $\tilde{\mathcal{O}}(\frac{1}{\sqrt{K}})$, with $K$ being the number of episodes in the offline data. In this work, we seek to understand instance-dependent bounds for offline RL with function approximation. We present an algorithm called Bootstrapped and Constrained Pessimistic Value Iteration (BCP-VI), which leverages data bootstrapping and constrained optimization on top of pessimism. We show that under a partial data coverage assumption, that of \emph{concentrability} with respect to an optimal policy, the proposed algorithm yields a fast rate of $\tilde{\mathcal{O}}(\frac{1}{K})$ for offline RL when there is a positive gap in the optimal Q-value functions, even when the offline data were adaptively collected. Moreover, when the linear features of the optimal actions in the states reachable by an optimal policy span those reachable by the behavior policy and the optimal actions are unique, offline RL achieves absolute zero sub-optimality error when $K$ exceeds a (finite) instance-dependent threshold. To the best of our knowledge, these are the first $\tilde{\mathcal{O}}(\frac{1}{K})$ bound and absolute zero sub-optimality bound respectively for offline RL with linear function approximation from adaptive data with partial coverage. We also provide instance-agnostic and instance-dependent information-theoretical lower bounds to complement our upper bounds.

cs.LG↗

arXiv2023

Who Should I Trust: AI or Myself? Leveraging Human and AI Correctness Likelihood to Promote Appropriate Trust in AI-Assisted Decision-Making

In AI-assisted decision-making, it is critical for human decision-makers to know when to trust AI and when to trust themselves. However, prior studies calibrated human trust only based on AI confidence indicating AI's correctness likelihood (CL) but ignored humans' CL, hindering optimal team decision-making. To mitigate this gap, we proposed to promote humans' appropriate trust based on the CL of both sides at a task-instance level. We first modeled humans' CL by approximating their decision-making models and computing their potential performance in similar instances. We demonstrated the feasibility and effectiveness of our model via two preliminary studies. Then, we proposed three CL exploitation strategies to calibrate users' trust explicitly/implicitly in the AI-assisted decision-making process. Results from a between-subjects experiment (N=293) showed that our CL exploitation strategies promoted more appropriate human trust in AI, compared with only using AI confidence. We further provided practical implications for more human-compatible AI-assisted decision-making.

cs.HC↗

arXiv2022

Offline Reinforcement Learning with Differentiable Function Approximation is Provably Efficient

Offline reinforcement learning, which aims at optimizing sequential decision-making strategies with historical data, has been extensively applied in real-life applications. State-Of-The-Art algorithms usually leverage powerful function approximators (e.g. neural networks) to alleviate the sample complexity hurdle for better empirical performances. Despite the successes, a more systematic understanding of the statistical complexity for function approximation remains lacking. Towards bridging the gap, we take a step by considering offline reinforcement learning with differentiable function class approximation (DFA). This function class naturally incorporates a wide range of models with nonlinear/nonconvex structures. Most importantly, we show offline RL with differentiable function approximation is provably efficient by analyzing the pessimistic fitted Q-learning (PFQL) algorithm, and our results provide the theoretical basis for understanding a variety of practical heuristics that rely on Fitted Q-Iteration style design. In addition, we further improve our guarantee with a tighter instance-dependent characterization. We hope our work could draw interest in studying reinforcement learning with differentiable function approximation beyond the scope of current research.

cs.LG↗

arXiv2022

Why Quantization Improves Generalization: NTK of Binary Weight Neural Networks

Quantized neural networks have drawn a lot of attention as they reduce the space and computational complexity during the inference. Moreover, there has been folklore that quantization acts as an implicit regularizer and thus can improve the generalizability of neural networks, yet no existing work formalizes this interesting folklore. In this paper, we take the binary weights in a neural network as random variables under stochastic rounding, and study the distribution propagation over different layers in the neural network. We propose a quasi neural network to approximate the distribution propagation, which is a neural network with continuous parameters and smooth activation function. We derive the neural tangent kernel (NTK) for this quasi neural network, and show that the eigenvalue of NTK decays at approximately exponential rate, which is comparable to that of Gaussian kernel with randomized scale. This in turn indicates that the Reproducing Kernel Hilbert Space (RKHS) of a binary weight neural network covers a strict subset of functions compared with the one with real value weights. We use experiments to verify that the quasi neural network we proposed can well approximate binary weight neural network. Furthermore, binary weight neural network gives a lower generalization gap compared with real value weight neural network, which is similar to the difference between Gaussian kernel and Laplace kernel.

cs.LG↗

arXiv2022

Offline Stochastic Shortest Path: Learning, Evaluation and Towards Optimality

Goal-oriented Reinforcement Learning, where the agent needs to reach the goal state while simultaneously minimizing the cost, has received significant attention in real-world applications. Its theoretical formulation, stochastic shortest path (SSP), has been intensively researched in the online setting. Nevertheless, it remains understudied when such an online interaction is prohibited and only historical data is provided. In this paper, we consider the offline stochastic shortest path problem when the state space and the action space are finite. We design the simple value iteration-based algorithms for tackling both offline policy evaluation (OPE) and offline policy learning tasks. Notably, our analysis of these simple algorithms yields strong instance-dependent bounds which can imply worst-case bounds that are near-minimax optimal. We hope our study could help illuminate the fundamental statistical limits of the offline SSP problem and motivate further studies beyond the scope of current consideration.

cs.LG↗

arXiv2022

Sample-Efficient Reinforcement Learning with loglog(T) Switching Cost

We study the problem of reinforcement learning (RL) with low (policy) switching cost - a problem well-motivated by real-life RL applications in which deployments of new policies are costly and the number of policy updates must be low. In this paper, we propose a new algorithm based on stage-wise exploration and adaptive policy elimination that achieves a regret of $\widetilde{O}(\sqrt{H^4S^2AT})$ while requiring a switching cost of $O(HSA \log\log T)$. This is an exponential improvement over the best-known switching cost $O(H^2SA\log T)$ among existing methods with $\widetilde{O}(\mathrm{poly}(H,S,A)\sqrt{T})$ regret. In the above, $S,A$ denotes the number of states and actions in an $H$-horizon episodic Markov Decision Process model with unknown transitions, and $T$ is the number of steps. As a byproduct of our new techniques, we also derive a reward-free exploration algorithm with a switching cost of $O(HSA)$. Furthermore, we prove a pair of information-theoretical lower bounds which say that (1) Any no-regret algorithm must have a switching cost of $Ω(HSA)$; (2) Any $\widetilde{O}(\sqrt{T})$ regret algorithm must incur a switching cost of $Ω(HSA\log\log T)$. Both our algorithms are thus optimal in their switching costs.

cs.LG↗

arXiv2022

A Holistic Framework for Analyzing the COVID-19 Vaccine Debate

The Covid-19 pandemic has led to infodemic of low quality information leading to poor health decisions. Combating the outcomes of this infodemic is not only a question of identifying false claims, but also reasoning about the decisions individuals make. In this work we propose a holistic analysis framework connecting stance and reason analysis, and fine-grained entity level moral sentiment analysis. We study how to model the dependencies between the different level of analysis and incorporate human insights into the learning process. Experiments show that our framework provides reliable predictions even in the low-supervision settings.

cs.CL↗

arXiv2022

Near-optimal Offline Reinforcement Learning with Linear Representation: Leveraging Variance Information with Pessimism

Offline reinforcement learning, which seeks to utilize offline/historical data to optimize sequential decision-making strategies, has gained surging prominence in recent studies. Due to the advantage that appropriate function approximators can help mitigate the sample complexity burden in modern reinforcement learning problems, existing endeavors usually enforce powerful function representation models (e.g. neural networks) to learn the optimal policies. However, a precise understanding of the statistical limits with function representations, remains elusive, even when such a representation is linear. Towards this goal, we study the statistical limits of offline reinforcement learning with linear model representations. To derive the tight offline learning bound, we design the variance-aware pessimistic value iteration (VAPVI), which adopts the conditional variance information of the value function for time-inhomogeneous episodic linear Markov decision processes (MDPs). VAPVI leverages estimated variances of the value functions to reweight the Bellman residuals in the least-square pessimistic value iteration and provides improved offline learning bounds over the best-known existing results (whereas the Bellman residuals are equally weighted by design). More importantly, our learning bounds are expressed in terms of system quantities, which provide natural instance-dependent characterizations that previous results are short of. We hope our results draw a clearer picture of what offline learning should look like when linear representations are provided.

cs.LG↗

arXiv2021

Towards Instance-Optimal Offline Reinforcement Learning with Pessimism

We study the offline reinforcement learning (offline RL) problem, where the goal is to learn a reward-maximizing policy in an unknown Markov Decision Process (MDP) using the data coming from a policy $μ$. In particular, we consider the sample complexity problems of offline RL for finite-horizon MDPs. Prior works study this problem based on different data-coverage assumptions, and their learning guarantees are expressed by the covering coefficients which lack the explicit characterization of system quantities. In this work, we analyze the Adaptive Pessimistic Value Iteration (APVI) algorithm and derive the suboptimality upper bound that nearly matches \[ O\left(\sum_{h=1}^H\sum_{s_h,a_h}d^{π^\star}_h(s_h,a_h)\sqrt{\frac{\mathrm{Var}_{P_{s_h,a_h}}{(V^\star_{h+1}+r_h)}}{d^μ_h(s_h,a_h)}}\sqrt{\frac{1}{n}}\right). \] In complementary, we also prove a per-instance information-theoretical lower bound under the weak assumption that $d^μ_h(s_h,a_h)>0$ if $d^{π^\star}_h(s_h,a_h)>0$. Different from the previous minimax lower bounds, the per-instance lower bound (via local minimaxity) is a much stronger criterion as it applies to individual instances separately. Here $π^\star$ is a optimal policy, $μ$ is the behavior policy and $d_h^μ$ is the marginal state-action probability. We call the above equation the intrinsic offline reinforcement learning bound since it directly implies all the existing optimal results: minimax rate under uniform data-coverage assumption, horizon-free setting, single policy concentrability, and the tight problem-dependent results. Later, we extend the result to the assumption-free regime (where we make no assumption on $ μ$) and obtain the assumption-free intrinsic bound. Due to its generic form, we believe the intrinsic bound could help illuminate what makes a specific problem hard and reveal the fundamental challenges in offline RL.

cs.LG↗

arXiv2021

Optimal Uniform OPE and Model-based Offline Reinforcement Learning in Time-Homogeneous, Reward-Free and Task-Agnostic Settings

This work studies the statistical limits of uniform convergence for offline policy evaluation (OPE) problems with model-based methods (for episodic MDP) and provides a unified framework towards optimal learning for several well-motivated offline tasks. Uniform OPE $\sup_Π|Q^π-\hat{Q}^π|<ε$ is a stronger measure than the point-wise OPE and ensures offline learning when $Π$ contains all policies (the global class). In this paper, we establish an $Ω(H^2 S/d_mε^2)$ lower bound (over model-based family) for the global uniform OPE and our main result establishes an upper bound of $\tilde{O}(H^2/d_mε^2)$ for the \emph{local} uniform convergence that applies to all \emph{near-empirically optimal} policies for the MDPs with \emph{stationary} transition. Here $d_m$ is the minimal marginal state-action probability. Critically, the highlight in achieving the optimal rate $\tilde{O}(H^2/d_mε^2)$ is our design of \emph{singleton absorbing MDP}, which is a new sharp analysis tool that works with the model-based approach. We generalize such a model-based framework to the new settings: offline task-agnostic and the offline reward-free with optimal complexity $\tilde{O}(H^2\log(K)/d_mε^2)$ ($K$ is the number of tasks) and $\tilde{O}(H^2S/d_mε^2)$ respectively. These results provide a unified solution for simultaneously solving different offline RL problems.

cs.LG↗

arXiv2021

Near-Optimal Offline Reinforcement Learning via Double Variance Reduction

We consider the problem of offline reinforcement learning (RL) -- a well-motivated setting of RL that aims at policy optimization using only historical data. Despite its wide applicability, theoretical understandings of offline RL, such as its optimal sample complexity, remain largely open even in basic settings such as \emph{tabular} Markov Decision Processes (MDPs). In this paper, we propose Off-Policy Double Variance Reduction (OPDVR), a new variance reduction based algorithm for offline RL. Our main result shows that OPDVR provably identifies an $ε$-optimal policy with $\widetilde{O}(H^2/d_mε^2)$ episodes of offline data in the finite-horizon stationary transition setting, where $H$ is the horizon length and $d_m$ is the minimal marginal state-action distribution induced by the behavior policy. This improves over the best known upper bound by a factor of $H$. Moreover, we establish an information-theoretic lower bound of $Ω(H^2/d_mε^2)$ which certifies that OPDVR is optimal up to logarithmic factors. Lastly, we show that OPDVR also achieves rate-optimal sample complexity under alternative settings such as the finite-horizon MDPs with non-stationary transitions and the infinite horizon MDPs with discounted rewards.

cs.LG↗

arXiv2020

Near-Optimal Provable Uniform Convergence in Offline Policy Evaluation for Reinforcement Learning

The problem of Offline Policy Evaluation (OPE) in Reinforcement Learning (RL) is a critical step towards applying RL in real-life applications. Existing work on OPE mostly focus on evaluating a fixed target policy $π$, which does not provide useful bounds for offline policy learning as $π$ will then be data-dependent. We address this problem by simultaneously evaluating all policies in a policy class $Π$ -- uniform convergence in OPE -- and obtain nearly optimal error bounds for a number of global / local policy classes. Our results imply that the model-based planning achieves an optimal episode complexity of $\widetilde{O}(H^3/d_mε^2)$ in identifying an $ε$-optimal policy under the time-inhomogeneous episodic MDP model ($H$ is the planning horizon, $d_m$ is a quantity that reflects the exploration of the logging policy $μ$). To the best of our knowledge, this is the first time the optimal rate is shown to be possible for the offline RL setting and the paper is the first that systematically investigates the uniform convergence in OPE.

cs.LG↗

arXiv2020

Asymptotically Efficient Off-Policy Evaluation for Tabular Reinforcement Learning

We consider the problem of off-policy evaluation for reinforcement learning, where the goal is to estimate the expected reward of a target policy $π$ using offline data collected by running a logging policy $μ$. Standard importance-sampling based approaches for this problem suffer from a variance that scales exponentially with time horizon $H$, which motivates a splurge of recent interest in alternatives that break the "Curse of Horizon" (Liu et al. 2018, Xie et al. 2019). In particular, it was shown that a marginalized importance sampling (MIS) approach can be used to achieve an estimation error of order $O(H^3/ n)$ in mean square error (MSE) under an episodic Markov Decision Process model with finite states and potentially infinite actions. The MSE bound however is still a factor of $H$ away from a Cramer-Rao lower bound of order $Ω(H^2/n)$. In this paper, we prove that with a simple modification to the MIS estimator, we can asymptotically attain the Cramer-Rao lower bound, provided that the action space is finite. We also provide a general method for constructing MIS estimators with high-probability error bounds.

cs.LG↗

arXiv2019

Shared Generative Latent Representation Learning for Multi-view Clustering

Clustering multi-view data has been a fundamental research topic in the computer vision community. It has been shown that a better accuracy can be achieved by integrating information of all the views than just using one view individually. However, the existing methods often struggle with the issues of dealing with the large-scale datasets and the poor performance in reconstructing samples. This paper proposes a novel multi-view clustering method by learning a shared generative latent representation that obeys a mixture of Gaussian distributions. The motivation is based on the fact that the multi-view data share a common latent embedding despite the diversity among the views. Specifically, benefited from the success of the deep generative learning, the proposed model not only can extract the nonlinear features from the views, but render a powerful ability in capturing the correlations among all the views. The extensive experimental results, on several datasets with different scales, demonstrate that the proposed method outperforms the state-of-the-art methods under a range of performance criteria.

cs.CV↗

arXiv2018

Working in Pairs: Understanding the Effects of Worker Interactions in Crowdwork

Crowdsourcing has gained popularity as a tool to harness human brain power to help solve problems that are difficult for computers. Previous work in crowdsourcing often assumes that workers complete crowdwork independently. In this paper, we relax the independent property of crowdwork and explore how introducing direct, synchronous, and free-style interactions between workers would affect crowdwork. In particular, motivated by the concept of peer instruction in educational settings, we study the effects of peer communication in crowdsourcing environments. In the crowdsourcing setting with peer communication, pairs of workers are asked to complete the same task together by first generating their initial answers to the task independently and then freely discussing the tasks with each other and updating their answers after the discussion. We experimentally examine the effects of peer communication in crowdwork on various common types of tasks on crowdsourcing platforms, including image labeling, optical character recognition (OCR), audio transcription, and nutrition analysis. Our experiment results show that the work quality is significantly improved in tasks with peer communication compared to tasks where workers complete the work independently. However, participating in tasks with peer communication has limited effects on influencing worker's independent performance in tasks of the same type in the future.

cs.HC↗

arXiv2018

Arago Optics: Maximal Confinement of Traveling Waves

Optics is limited in the 'ray-approximation'-inclusion of wave properties result in additional phenomena and applications; interferometers and diffraction gratings are two manifestations of such non-geometric, physical optics. Incidentally, the most precise measurement ever, at one part per 10^21 in the (2017) Nobel winning discovery of gravitational waves was achieved with an interferometer. Amendments to the properties of the medium promise negative refractive index meta-materials, perfect imaging, light cloaking, and other ultra-natural marvels. Attention to photon phase, correlations, statistics and wavelength independent phase shifts result in singular optics, quantum optics and anholonomy. Here we present another possibility, namely 'Arago-optics' to maximize the efficacy of a device by strategically deploying the key qualities along its perimeter. For instance, in conventional sources, waves are generated with maximum intensity at the core; whereas in an Arago-source, intensity is minimal or zero at the center, but highest on villus stretches at the margins. We reason that for a given size and energy output, this radiation profile, produces the highest concentration of energy at the focus, with the maximal confinement of the wave packet. Likewise, the utmost detector resolution is attained when sensitivity is highest on the perimeter and less at the center. This concept holds beyond ultra-focus and Gaussian beams, but generally applies to beams of 'waves' that show constructive and destructive interference. The idea is particularly well suited for a fresh integration of geometry and topology with electronics and materials into real-time wave engineering.

physics.gen-ph↗