Ruosong Wang

NeurIPS Conference 2025 Conference Paper

Deployment Efficient Reward-Free Exploration with Linear Function Approximation

• Zihan Zhang
• Yuxin Chen
• Jason Lee
• Simon Du
• Lin Yang
• Ruosong Wang

We study deployment-efficient reward-free exploration with linear function approximation, where the goal is to explore a linear Markov Decision Process (MDP) without revealing the reward function, while minimizing the number of distinct policies implemented during learning. By ``deployment efficient'', we mean algorithms that require few policies deployed during exploration -- crucial in real-world applications where such deployments are costly or disruptive. We design a novel reinforcement learning algorithm that achieves near-optimal deployment efficiency for linear MDPs in the reward-free setting, using at most $H$ exploration policies during execution (where $H$ is the horizon length), while maintaining sample complexity polynomial in feature dimension and horizon length. Unlike previous approaches with similar deployment efficiency guarantees, our algorithm's sample complexity is independent of the reachability or explorability coefficients of the underlying MDP, which can be arbitrarily small and lead to unbounded sample complexity in certain cases -- directly addressing an open problem from prior work. Our technical contributions include a data-dependent method for truncating state-action pairs in linear MDPs, efficient offline policy evaluation and optimization algorithms for these truncated MDPs, and a careful integration of these components to implement reward-free exploration with linear function approximation without sacrificing deployment efficiency.

ICML Conference 2025 Conference Paper

Minimax Optimal Regret Bound for Reinforcement Learning with Trajectory Feedback

• Zihan Zhang
• Yuxin Chen 0002
• Jason D. Lee
• Simon S. Du
• Ruosong Wang

In this work, we study reinforcement learning (RL) with trajectory feedback. Compared to the standard RL setting, in RL with trajectory feedback, the agent only observes the accumulative reward along the trajectory, and therefore, this model is particularly suitable for scenarios where querying the reward in each single step incurs prohibitive cost. For a finite-horizon Markov Decision Process (MDP) with $S$ states, $A$ actions and a horizon length of $H$, we develop an algorithm that enjoys an asymptotically nearly optimal regret of $\tilde{O}\left(\sqrt{SAH^3K}\right)$ in $K$ episodes. To achieve this result, our new technical ingredients include (i) constructing a tighter confidence region for the reward function by incorporating the RL with trajectory feedback setting with techniques in linear bandits and (ii) constructing a reference transition model to better guide the exploration process.

ICLR Conference 2025 Conference Paper

Misspecified Q-Learning with Sparse Linear Function Approximation: Tight Bounds on Approximation Error

• Ally Yalei Du
• Lin F. Yang
• Ruosong Wang

The recent work by Dong and Yang (2023) showed for misspecified sparse linear bandits, one can obtain an $O(\epsilon)$-optimal policy using a polynomial number of samples when the sparsity is a constant, where $\epsilon$ is the misspecification error. This result is in sharp contrast to misspecified linear bandits without sparsity, which require an exponential number of samples to get the same guarantee. In order to study whether the analog result is possible in the reinforcement learning setting, we consider the following problem: assuming the optimal $Q$-function is a $d$-dimensional linear function with sparsity $k$ and misspecification error $\epsilon$, whether we can obtain an $O(\epsilon)$-optimal policy using number of samples polynomially in the feature dimension $d$. We first demonstrate why the standard approach based on Bellman backup or the existing optimistic value function elimination approach such as OLIVE (Jiang et al., 2017) achieves suboptimal guarantees for this problem. We then design a novel elimination-based algorithm to show one can obtain an $O(H\epsilon)$-optimal policy with sample complexity polynomially in the feature dimension $d$ and planning horizon $H$. Lastly, we complement our upper bound with an $\tilde \Omega(H\epsilon)$ suboptimality lower bound, giving a complete picture of this problem.

ICLR Conference 2025 Conference Paper

Regret-Optimal List Replicable Bandit Learning: Matching Upper and Lower Bounds

• Michael Chen
• Aduri Pavan
• N. V. Vinodchandran
• Ruosong Wang
• Lin F. Yang

This paper investigates *list replicability* [Dixon et al., 2023] in the context of multi-armed (also linear) bandits (MAB). We define an algorithm $A$ for MAB to be $(\ell,\delta)$-list replicable if with probability at least $1-\delta$, $A$ has at most $\ell$ traces in independent executions even with different random bits, where a trace means sequence of arms played during an execution. For $k$-armed bandits, although the total number of traces can be $\Omega(k^T)$ for a time horizon $T$, we present several surprising upper bounds that either independent of or logarithmic of $T$: (1) a $(2^{k},\delta)$-list replicable algorithm with near-optimal regret, $\widetilde{O}({\sqrt{kT}})$, (2) a $(O(k/\delta),\delta)$-list replicable algorithm with regret $\widetilde{O}\left(\frac{k}{\delta}\sqrt{kT}\right)$, (3) a $((k+1)^{B-1}, \delta)$-list replicable algorithm with regret $\widetilde{O}(k^{\frac{3}{2}}T^{{\frac{1}{2}}+2^{-(B+1)}})$ for any integer $B>1$. On the other hand, for the sublinear regret regime, we establish a matching lower bound on the list complexity (parameter $\ell$). We prove that there is no $(k-1,\delta)$-list replicable algorithm with $o(T)$-regret. This is optimal in list complexity in the sub-linear regret regime as there is a $(k, 0)$-list replicable algorithm with $O(T^{2/3})$-regret. We further show that for linear bandits with $d$-dimensional features, there is a $\widetilde{O}(d^2T^{1/2+2^{-(B+1)}})$-regret algorithm with $((2d+1)^{B-1},\delta)$-list replicability, for $B>1$, even when the number of possible arms can be infinite.

NeurIPS Conference 2024 Conference Paper

Uniform Last-Iterate Guarantee for Bandits and Reinforcement Learning

• Junyan Liu
• Yunfan Li
• Ruosong Wang
• Lin F. Yang

Existing metrics for reinforcement learning (RL) such as regret, PAC bounds, or uniform-PAC (Dann et al. , 2017), typically evaluate the cumulative performance, while allowing the play of an arbitrarily bad policy at any finite time t. Such a behavior can be highly detrimental in high-stakes applications. This paper introduces a stronger metric, uniform last-iterate (ULI) guarantee, capturing both cumulative and instantaneous performance of RL algorithms. Specifically, ULI characterizes the instantaneous performance since it ensures that the per-round suboptimality of the played policy is bounded by a function, monotonically decreasing w. r. t. (large) round t, preventing revisits to bad policies when sufficient samples are available. We demonstrate that a near-optimal ULI guarantee directly implies near-optimal cumulative performance across aforementioned metrics, but not the other way around. To examine the achievability of ULI, we first provide two positive results for bandit problems with finite arms, showing that some elimination-based algorithms and high-probability adversarial algorithms with stronger analysis or additional designs, can attain near-optimal ULI guarantees. We also provide a negative result, indicating that optimistic algorithms cannot achieve a near-optimal ULI guarantee. Furthermore, we propose an efficient algorithm for linear bandits with infinitely many arms, which achieves the ULI guarantee, given access to an optimization oracle. Finally, we propose an algorithm that achieves a near-optimal ULI guarantee for the online reinforcement learning setting.

ICML Conference 2023 Conference Paper

Horizon-Free and Variance-Dependent Reinforcement Learning for Latent Markov Decision Processes

• Runlong Zhou
• Ruosong Wang
• Simon S. Du

We study regret minimization for reinforcement learning (RL) in Latent Markov Decision Processes (LMDPs) with context in hindsight. We design a novel model-based algorithmic framework which can be instantiated with both a model-optimistic and a value-optimistic solver. We prove an $\tilde{O}(\sqrt{\mathsf{Var}^\star M \Gamma S A K})$ regret bound where $\tilde{O}$ hides logarithm factors, $M$ is the number of contexts, $S$ is the number of states, $A$ is the number of actions, $K$ is the number of episodes, $\Gamma \le S$ is the maximum transition degree of any state-action pair, and $\mathsf{Var}^\star$ is a variance quantity describing the determinism of the LMDP. The regret bound only scales logarithmically with the planning horizon, thus yielding the first (nearly) horizon-free regret bound for LMDP. This is also the first problem-dependent regret bound for LMDP. Key in our proof is an analysis of the total variance of alpha vectors (a generalization of value functions), which is handled with a truncation method. We complement our positive result with a novel $\Omega(\sqrt{\mathsf{Var}^\star M S A K})$ regret lower bound with $\Gamma = 2$, which shows our upper bound minimax optimal when $\Gamma$ is a constant for the class of variance-bounded LMDPs. Our lower bound relies on new constructions of hard instances and an argument inspired by the symmetrization technique from theoretical computer science, both of which are technically different from existing lower bound proof for MDPs, and thus can be of independent interest.

ICLR Conference 2023 Conference Paper

Variance-Aware Sparse Linear Bandits

• Yan Dai 0002
• Ruosong Wang
• Simon S. Du

It is well-known that for sparse linear bandits, when ignoring the dependency on sparsity which is much smaller than the ambient dimension, the worst-case minimax regret is $\widetilde{\Theta}\left(\sqrt{dT}\right)$ where $d$ is the ambient dimension and $T$ is the number of rounds. On the other hand, in the benign setting where there is no noise and the action set is the unit sphere, one can use divide-and-conquer to achieve $\widetilde{\mathcal O}(1)$ regret, which is (nearly) independent of $d$ and $T$. In this paper, we present the first variance-aware regret guarantee for sparse linear bandits: $\widetilde{\mathcal O}\left(\sqrt{d\sum_{t=1}^T \sigma_t^2} + 1\right)$, where $\sigma_t^2$ is the variance of the noise at the $t$-th round. This bound naturally interpolates the regret bounds for the worst-case constant-variance regime (i.e., $\sigma_t \equiv \Omega(1)$) and the benign deterministic regimes (i.e., $\sigma_t \equiv 0$). To achieve this variance-aware regret guarantee, we develop a general framework that converts any variance-aware linear bandit algorithm to a variance-aware algorithm for sparse linear bandits in a "black-box" manner. Specifically, we take two recent algorithms as black boxes to illustrate that the claimed bounds indeed hold, where the first algorithm can handle unknown-variance cases and the second one is more efficient.

NeurIPS Conference 2021 Conference Paper

An Exponential Lower Bound for Linearly Realizable MDP with Constant Suboptimality Gap

• Yuanhao Wang
• Ruosong Wang
• Sham Kakade

A fundamental question in the theory of reinforcement learning is: suppose the optimal $Q$-function lies in the linear span of a given $d$ dimensional feature mapping, is sample-efficient reinforcement learning (RL) possible? The recent and remarkable result of Weisz et al. (2020) resolves this question in the negative, providing an exponential (in $d$) sample size lower bound, which holds even if the agent has access to a generative model of the environment. One may hope that such a lower can be circumvented with an even stronger assumption that there is a \emph{constant gap} between the optimal $Q$-value of the best action and that of the second-best action (for all states); indeed, the construction in Weisz et al. (2020) relies on having an exponentially small gap. This work resolves this subsequent question, showing that an exponential sample complexity lower bound still holds even if a constant gap is assumed. Perhaps surprisingly, this result implies an exponential separation between the online RL setting and the generative model setting, where sample-efficient RL is in fact possible in the latter setting with a constant gap. Complementing our negative hardness result, we give two positive results showing that provably sample-efficient RL is possible either under an additional low-variance assumption or under a novel hypercontractivity assumption.

ICML Conference 2021 Conference Paper

Bilinear Classes: A Structural Framework for Provable Generalization in RL

• Simon S. Du
• Sham M. Kakade
• Jason D. Lee
• Shachar Lovett
• Gaurav Mahajan
• Wen Sun 0002
• Ruosong Wang

This work introduces Bilinear Classes, a new structural framework, which permit generalization in reinforcement learning in a wide variety of settings through the use of function approximation. The framework incorporates nearly all existing models in which a polynomial sample complexity is achievable, and, notably, also includes new models, such as the Linear Q*/V* model in which both the optimal Q-function and the optimal V-function are linear in some known feature space. Our main result provides an RL algorithm which has polynomial sample complexity for Bilinear Classes; notably, this sample complexity is stated in terms of a reduction to the generalization error of an underlying supervised learning sub-problem. These bounds nearly match the best known sample complexity bounds for existing models. Furthermore, this framework also extends to the infinite dimensional (RKHS) setting: for the the Linear Q*/V* model, linear MDPs, and linear mixture MDPs, we provide sample complexities that have no explicit dependence on the explicit feature dimension (which could be infinite), but instead depends only on information theoretic quantities.

ICML Conference 2021 Conference Paper

Instabilities of Offline RL with Pre-Trained Neural Representation

• Ruosong Wang
• Yifan Wu
• Ruslan Salakhutdinov
• Sham M. Kakade

In offline reinforcement learning (RL), we seek to utilize offline data to evaluate (or learn) policies in scenarios where the data are collected from a distribution that substantially differs from that of the target policy to be evaluated. Recent theoretical advances have shown that such sample-efficient offline RL is indeed possible provided certain strong representational conditions hold, else there are lower bounds exhibiting exponential error amplification (in the problem horizon) unless the data collection distribution has only a mild distribution shift relative to the target policy. This work studies these issues from an empirical perspective to gauge how stable offline RL methods are. In particular, our methodology explores these ideas when using features from pre-trained neural networks, in the hope that these representations are powerful enough to permit sample efficient offline RL. Through extensive experiments on a range of tasks, we see that substantial error amplification does occur even when using such pre-trained representations (trained on the same task itself); we find offline RL is stable only under extremely mild distribution shift. The implications of these results, both from a theoretical and an empirical perspective, are that successful offline RL (where we seek to go beyond the low distribution shift regime) requires substantially stronger conditions beyond those which suffice for successful supervised learning.

ICLR Conference 2021 Conference Paper

Optimism in Reinforcement Learning with Generalized Linear Function Approximation

• Yining Wang 0001
• Ruosong Wang
• Simon S. Du
• Akshay Krishnamurthy

We design a new provably efficient algorithm for episodic reinforcement learning with generalized linear function approximation. We analyze the algorithm under a new expressivity assumption that we call ``optimistic closure,'' which is strictly weaker than assumptions from prior analyses for the linear setting. With optimistic closure, we prove that our algorithm enjoys a regret bound of $\widetilde{O}\left(H\sqrt{d^3 T}\right)$ where $H$ is the horizon, $d$ is the dimensionality of the state-action features and $T$ is the number of episodes. This is the first statistically and computationally efficient algorithm for reinforcement learning with generalized linear functions.

FOCS Conference 2021 Conference Paper

Settling the Horizon-Dependence of Sample Complexity in Reinforcement Learning

• Yuanzhi Li
• Ruosong Wang
• Lin F. Yang

Recently there is a surge of interest in under-standing the horizon-dependence of the sample complexity in reinforcement learning (RL). Notably, for an RL environment with horizon length H, previous work have shown that there is a probably approximately correct (PAC) algorithm that learns an O(1)-optimal policy using polylog(H) episodes of environment interactions when the number of states and actions is fixed. It is yet unknown whether the polylog (H) dependence is necessary or not. In this work, we resolve this question by developing an algorithm that achieves the same PAC guarantee while using only O(1) episodes of environment interactions, completely settling the horizon-dependence of the sample complexity in RL. We achieve this bound by (i) establishing a connection between value functions in discounted and finite-horizon Markov decision processes (MDPs) and (ii) a novel perturbation analysis in MDPs. We believe our new techniques are of independent interest and could be applied in related questions in RL.

ICLR Conference 2021 Conference Paper

What are the Statistical Limits of Offline RL with Linear Function Approximation?

• Ruosong Wang
• Dean P. Foster
• Sham M. Kakade

Offline reinforcement learning seeks to utilize offline (observational) data to guide the learning of (causal) sequential decision making strategies. The hope is that offline reinforcement learning coupled with function approximation methods (to deal with the curse of dimensionality) can provide a means to help alleviate the excessive sample complexity burden in modern sequential decision making problems. However, the extent to which this broader approach can be effective is not well understood, where the literature largely consists of sufficient conditions. This work focuses on the basic question of what are necessary representational and distributional conditions that permit provable sample-efficient offline reinforcement learning. Perhaps surprisingly, our main result shows that even if: i) we have realizability in that the true value function of \emph{every} policy is linear in a given set of features and 2) our off-policy data has good coverage over all features (under a strong spectral condition), any algorithm still (information-theoretically) requires a number of offline samples that is exponential in the problem horizon to non-trivially estimate the value of \emph{any} given policy. Our results highlight that sample-efficient offline policy evaluation is not possible unless significantly stronger conditions hold; such conditions include either having low distribution shift (where the offline data distribution is close to the distribution of the policy to be evaluated) or significantly stronger representational conditions (beyond realizability).

NeurIPS Conference 2020 Conference Paper

Agnostic $Q$-learning with Function Approximation in Deterministic Systems: Near-Optimal Bounds on Approximation Error and Sample Complexity

• Simon S. Du
• Jason D. Lee
• Gaurav Mahajan
• Ruosong Wang

The current paper studies the problem of agnostic $Q$-learning with function approximation in deterministic systems where the optimal $Q$-function is approximable by a function in the class $\mathcal{F}$ with approximation error $\delta \ge 0$. We propose a novel recursion-based algorithm and show that if $\delta = O\left(\rho/\sqrt{\dim_E}\right)$, then one can find the optimal policy using $O(\dim_E)$ trajectories, where $\rho$ is the gap between the optimal $Q$-value of the best actions and that of the second-best actions and $\dim_E$ is the Eluder dimension of $\mathcal{F}$. Our result has two implications: \begin{enumerate} \item In conjunction with the lower bound in [Du et al. , 2020], our upper bound suggests that the condition $\delta = \widetilde{\Theta}\left(\rho/\sqrt{\dim_E}\right)$ is necessary and sufficient for algorithms with polynomial sample complexity. \item In conjunction with the obvious lower bound in the tabular case, our upper bound suggests that the sample complexity $\widetilde{\Theta}\left(\dim_E\right)$ is tight in the agnostic setting. \end{enumerate} Therefore, we help address the open problem on agnostic $Q$-learning proposed in [Wen and Van Roy, 2013]. We further extend our algorithm to the stochastic reward setting and obtain similar results.

ICLR Conference 2020 Conference Paper

Harnessing the Power of Infinitely Wide Deep Nets on Small-data Tasks

• Sanjeev Arora
• Simon S. Du
• Zhiyuan Li 0005
• Ruslan Salakhutdinov
• Ruosong Wang
• Dingli Yu

Recent research shows that the following two models are equivalent: (a) infinitely wide neural networks (NNs) trained under l2 loss by gradient descent with infinitesimally small learning rate (b) kernel regression with respect to so-called Neural Tangent Kernels (NTKs) (Jacot et al., 2018). An efficient algorithm to compute the NTK, as well as its convolutional counterparts, appears in Arora et al. (2019a), which allowed studying performance of infinitely wide nets on datasets like CIFAR-10. However, super-quadratic running time of kernel methods makes them best suited for small-data tasks. We report results suggesting neural tangent kernels perform strongly on low-data tasks. 1. On a standard testbed of classification/regression tasks from the UCI database, NTK SVM beats the previous gold standard, Random Forests (RF), and also the corresponding finite nets. 2. On CIFAR-10 with 10 – 640 training samples, Convolutional NTK consistently beats ResNet-34 by 1% - 3%. 3. On VOC07 testbed for few-shot image classification tasks on ImageNet with transfer learning (Goyal et al., 2019), replacing the linear SVM currently used with a Convolutional NTK SVM consistently improves performance. 4. Comparing the performance of NTK with the finite-width net it was derived from, NTK behavior starts at lower net widths than suggested by theoretical analysis(Arora et al., 2019a). NTK’s efficacy may trace to lower variance of output.

ICLR Conference 2020 Conference Paper

Is a Good Representation Sufficient for Sample Efficient Reinforcement Learning?

• Simon S. Du
• Sham M. Kakade
• Ruosong Wang
• Lin F. Yang

Modern deep learning methods provide effective means to learn good representations. However, is a good representation itself sufficient for sample efficient reinforcement learning? This question has largely been studied only with respect to (worst-case) approximation error, in the more classical approximate dynamic programming literature. With regards to the statistical viewpoint, this question is largely unexplored, and the extant body of literature mainly focuses on conditions which \emph{permit} sample efficient reinforcement learning with little understanding of what are \emph{necessary} conditions for efficient reinforcement learning. This work shows that, from the statistical viewpoint, the situation is far subtler than suggested by the more traditional approximation viewpoint, where the requirements on the representation that suffice for sample efficient RL are even more stringent. Our main results provide sharp thresholds for reinforcement learning methods, showing that there are hard limitations on what constitutes good function approximation (in terms of the dimensionality of the representation), where we focus on natural representational conditions relevant to value-based, model-based, and policy-based learning. These lower bounds highlight that having a good (value-based, model-based, or policy-based) representation in and of itself is insufficient for efficient reinforcement learning, unless the quality of this approximation passes certain hard thresholds. Furthermore, our lower bounds also imply exponential separations on the sample complexity between 1) value-based learning with perfect representation and value-based learning with a good-but-not-perfect representation, 2) value-based learning and policy-based learning, 3) policy-based learning and supervised learning and 4) reinforcement learning and imitation learning.

NeurIPS Conference 2020 Conference Paper

Is Long Horizon RL More Difficult Than Short Horizon RL?

• Ruosong Wang
• Simon S. Du
• Lin Yang
• Sham Kakade

Learning to plan for long horizons is a central challenge in episodic reinforcement learning problems. A fundamental question is to understand how the difficulty of the problem scales as the horizon increases. Here the natural measure of sample complexity is a normalized one: we are interested in the \emph{number of episodes} it takes to provably discover a policy whose value is $\varepsilon$ near to that of the optimal value, where the value is measured by the \emph{normalized} cumulative reward in each episode. In a COLT 2018 open problem, Jiang and Agarwal conjectured that, for tabular, episodic reinforcement learning problems, there exists a sample complexity lower bound which exhibits a polynomial dependence on the horizon --- a conjecture which is consistent with all known sample complexity upper bounds. This work refutes this conjecture, proving that tabular, episodic reinforcement learning is possible with a sample complexity that scales only \emph{logarithmically} with the planning horizon. In other words, when the values are appropriately normalized (to lie in the unit interval), this results shows that long horizon RL is no more difficult than short horizon RL, at least in a minimax sense. Our analysis introduces two ideas: (i) the construction of an $\varepsilon$-net for near-optimal policies whose log-covering number scales only logarithmically with the planning horizon, and (ii) the Online Trajectory Synthesis algorithm, which adaptively evaluates all policies in a given policy class and enjoys a sample complexity that scales logarithmically with the cardinality of the given policy class. Both may be of independent interest.

ICML Conference 2020 Conference Paper

Nearly Linear Row Sampling Algorithm for Quantile Regression

• Yi Li 0002
• Ruosong Wang
• Lin F. Yang
• Hanrui Zhang 0001

We give a row sampling algorithm for the quantile loss function with sample complexity nearly linear in the dimensionality of the data, improving upon the previous best algorithm whose sampling complexity has at least cubic dependence on the dimensionality. Based upon our row sampling algorithm, we give the fastest known algorithm for quantile regression and a graph sparsification algorithm for balanced directed graphs. Our main technical contribution is to show that Lewis weights sampling, which has been used in row sampling algorithms for $\ell_p$ norms, can also be applied in row sampling algorithms for a variety of loss functions. We complement our theoretical results by experiments to demonstrate the practicality of our approach.

NeurIPS Conference 2020 Conference Paper

On Reward-Free Reinforcement Learning with Linear Function Approximation

• Ruosong Wang
• Simon S. Du
• Lin Yang
• Russ R. Salakhutdinov

Reward-free reinforcement learning (RL) is a framework which is suitable for both the batch RL setting and the setting where there are many reward functions of interest. During the exploration phase, an agent collects samples without using a pre-specified reward function. After the exploration phase, a reward function is given, and the agent uses samples collected during the exploration phase to compute a near-optimal policy. Jin et al. [2020] showed that in the tabular setting, the agent only needs to collect polynomial number of samples (in terms of the number states, the number of actions, and the planning horizon) for reward-free RL. However, in practice, the number of states and actions can be large, and thus function approximation schemes are required for generalization. In this work, we give both positive and negative results for reward-free RL with linear function approximation. We give an algorithm for reward-free RL in the linear Markov decision process setting where both the transition and the reward admit linear representations. The sample complexity of our algorithm is polynomial in the feature dimension and the planning horizon, and is completely independent of the number of states and actions. We further give an exponential lower bound for reward-free RL in the setting where only the optimal $Q$-function admits a linear representation. Our results imply several interesting exponential separations on the sample complexity of reward-free RL.

NeurIPS Conference 2020 Conference Paper

Planning with General Objective Functions: Going Beyond Total Rewards

• Ruosong Wang
• Peilin Zhong
• Simon S. Du
• Russ R. Salakhutdinov
• Lin Yang

Standard sequential decision-making paradigms aim to maximize the cumulative reward when interacting with the unknown environment. , i. e. , maximize $\sum_{h = 1}^H r_h$ where $H$ is the planning horizon. However, this paradigm fails to model important practical applications, e. g. , safe control that aims to maximize the lowest reward, i. e. , maximize $\min_{h= 1}^H r_h$. In this paper, based on techniques in sketching algorithms, we propose a novel planning algorithm in deterministic systems which deals with a large class of objective functions of the form $f(r_1, r_2, .. . r_H)$ that are of interest to practical applications. We show that efficient planning is possible if $f$ is symmetric under permutation of coordinates and satisfies certain technical conditions. Complementing our algorithm, we further prove that removing any of the conditions will make the problem intractable in the worst case and thus demonstrate the necessity of our conditions.

NeurIPS Conference 2020 Conference Paper

Preference-based Reinforcement Learning with Finite-Time Guarantees

• Yichong Xu
• Ruosong Wang
• Lin Yang
• Aarti Singh
• Artur Dubrawski

Preference-based Reinforcement Learning (PbRL) replaces reward values in traditional reinforcement learning by preferences to better elicit human opinion on the target objective, especially when numerical reward values are hard to design or interpret. Despite promising results in applications, the theoretical understanding of PbRL is still in its infancy. In this paper, we present the first finite-time analysis for general PbRL problems. We first show that a unique optimal policy may not exist if preferences over trajectories are deterministic for PbRL. If preferences are stochastic, and the preference probability relates to the hidden reward values, we present algorithms for PbRL, both with and without a simulator, that are able to identify the best policy up to accuracy $\varepsilon$ with high probability. Our method explores the state space by navigating to under-explored states, and solves PbRL using a combination of dueling bandits and policy search. Experiments show the efficacy of our method when it is applied to real-world problems.

NeurIPS Conference 2020 Conference Paper

Provably Efficient Exploration for Reinforcement Learning Using Unsupervised Learning

• Fei Feng
• Ruosong Wang
• Wotao Yin
• Simon S. Du
• Lin Yang

Motivated by the prevailing paradigm of using unsupervised learning for efficient exploration in reinforcement learning (RL) problems [tang2017exploration, bellemare2016unifying], we investigate when this paradigm is provably efficient. We study episodic Markov decision processes with rich observations generated from a small number of latent states. We present a general algorithmic framework that is built upon two components: an unsupervised learning algorithm and a no-regret tabular RL algorithm. Theoretically, we prove that as long as the unsupervised learning algorithm enjoys a polynomial sample complexity guarantee, we can find a near-optimal policy with sample complexity polynomial in the number of latent states, which is significantly smaller than the number of observations. Empirically, we instantiate our framework on a class of hard exploration problems to demonstrate the practicality of our theory.

NeurIPS Conference 2020 Conference Paper

Reinforcement Learning with General Value Function Approximation: Provably Efficient Approach via Bounded Eluder Dimension

• Ruosong Wang
• Russ R. Salakhutdinov
• Lin Yang

Value function approximation has demonstrated phenomenal empirical success in reinforcement learning (RL). Nevertheless, despite a handful of recent progress on developing theory for RL with linear function approximation, the understanding of \emph{general} function approximation schemes largely remains missing. In this paper, we establish the first provably efficient RL algorithm with general value function approximation. We show that if the value functions admit an approximation with a function class $\mathcal{F}$, our algorithm achieves a regret bound of $\widetilde{O}(\mathrm{poly}(dH)\sqrt{T})$ where $d$ is a complexity measure of $\mathcal{F}$ that depends on the eluder dimension~[Russo and Van Roy, 2013] and log-covering numbers, $H$ is the planning horizon, and $T$ is the number interactions with the environment. Our theory generalizes the linear MDP assumption to general function classes. Moreover, our algorithm is model-free and provides a framework to justify the effectiveness of algorithms used in practice.

SODA Conference 2020 Conference Paper

The Communication Complexity of Optimization

• Santosh S. Vempala
• Ruosong Wang
• David P. Woodruff

SODA Conference 2020 Conference Paper

Tight Bounds for the Subspace Sketch Problem with Applications

• Yi Li 0002
• Ruosong Wang
• David P. Woodruff

In the subspace sketch problem one is given an n × d matrix A with O (log( nd )) bit entries, and would like to compress it in an arbitrary way to build a small space data structure Q p, so that for any given x ϵ ℝ d, with probability at least 2/3, one has Q p ( x ) = (1 ± ε )|| Ax || p, where p ≥ 0 and the randomness is over the construction of Q p. The central question is: How many bits are necessary to store Q p? This problem has applications to the communication of approximating the number of non-zeros in a matrix product, the size of coresets in projective clustering, the memory of streaming algorithms for regression in the row-update model, and embedding subspaces of L p in functional analysis. A major open question is the dependence on the approximation factor ε. We show if p ≥ 0 is not a positive even integer and d = Ω(log(1/ ε )), then ( ε −2 · d ) bits are necessary. On the other hand, if p is a positive even integer, then there is an upper bound of O ( d p log( nd )) bits independent of ε. Our results are optimal up to logarithmic factors, and show in particular that one cannot compress A to O ( d ) “directions” ν 1, …, ν o ( d ), such that for any x, || Ax || 1 can be well-approximated from 〈 ν 1, x 〉, …, 〈 ν O ( d ), x 〉. Our lower bound rules out arbitrary functions of these inner products (and in fact arbitrary data structures built from A ), and thus rules out the possibility of a singular value decomposition for ℓ 1 in a very strong sense. Indeed, as ε → 0, for p = 1 the space complexity becomes arbitrarily large, while for p = 2 it is at most O ( d 2 log( nd )). As corollaries of our main lower bound, we obtain new lower bounds for a wide range of applications, including the above, which in many cases are optimal.

ICML Conference 2019 Conference Paper

Dimensionality Reduction for Tukey Regression

• Kenneth L. Clarkson
• Ruosong Wang
• David P. Woodruff

We give the first dimensionality reduction methods for the overconstrained Tukey regression problem. The Tukey loss function $\|y\|_M = \sum_i M(y_i)$ has $M(y_i) \approx |y_i|^p$ for residual errors $y_i$ smaller than a prescribed threshold $\tau$, but $M(y_i)$ becomes constant for errors $|y_i| > \tau$. Our results depend on a new structural result, proven constructively, showing that for any $d$-dimensional subspace $L \subset \mathbb{R}^n$, there is a fixed bounded-size subset of coordinates containing, for every $y \in L$, all the large coordinates, with respect to the Tukey loss function, of $y$. Our methods reduce a given Tukey regression problem to a smaller weighted version, whose solution is a provably good approximate solution to the original problem. Our reductions are fast, simple and easy to implement, and we give empirical results demonstrating their practicality, using existing heuristic solvers for the small versions. We also give exponential-time algorithms giving provably good solutions, and hardness results suggesting that a significant speedup in the worst case is unlikely.

NeurIPS Conference 2019 Conference Paper

Efficient Symmetric Norm Regression via Linear Sketching

• Zhao Song
• Ruosong Wang
• Lin Yang
• Hongyang Zhang
• Peilin Zhong

We provide efficient algorithms for overconstrained linear regression problems with size $n \times d$ when the loss function is a symmetric norm (a norm invariant under sign-flips and coordinate-permutations). An important class of symmetric norms are Orlicz norms, where for a function $G$ and a vector $y \in \mathbb{R}^n$, the corresponding Orlicz norm $\|y\|_G$ is defined as the unique value $\alpha$ such that $\sum_{i=1}^n G(|y_i|/\alpha) = 1$. When the loss function is an Orlicz norm, our algorithm produces a $(1 + \varepsilon)$-approximate solution for an arbitrarily small constant $\varepsilon > 0$ in input-sparsity time, improving over the previously best-known algorithm which produces a $d \cdot \polylog n$-approximate solution. When the loss function is a general symmetric norm, our algorithm produces a $\sqrt{d} \cdot \polylog n \cdot \mathrm{mmc}(\ell)$-approximate solution in input-sparsity time, where $\mathrm{mmc}(\ell)$ is a quantity related to the symmetric norm under consideration. To the best of our knowledge, this is the first input-sparsity time algorithm with provable guarantees for the general class of symmetric norm regression problem. Our results shed light on resolving the universal sketching problem for linear regression, and the techniques might be of independent interest to numerical linear algebra problems more broadly.

ICML Conference 2019 Conference Paper

Fine-Grained Analysis of Optimization and Generalization for Overparameterized Two-Layer Neural Networks

• Sanjeev Arora
• Simon S. Du
• Wei Hu 0014
• Zhiyuan Li 0005
• Ruosong Wang

Recent works have cast some light on the mystery of why deep nets fit any data and generalize despite being very overparametrized. This paper analyzes training and generalization for a simple 2-layer ReLU net with random initialization, and provides the following improvements over recent works: (i) Using a tighter characterization of training speed than recent papers, an explanation for why training a neural net with random labels leads to slower training, as originally observed in [Zhang et al. ICLR’17]. (ii) Generalization bound independent of network size, using a data-dependent complexity measure. Our measure distinguishes clearly between random labels and true labels on MNIST and CIFAR, as shown by experiments. Moreover, recent papers require sample complexity to increase (slowly) with the size, while our sample complexity is completely independent of the network size. (iii) Learnability of a broad class of smooth functions by 2-layer ReLU nets trained via gradient descent. The key idea is to track dynamics of training and generalization via properties of a related kernel.

NeurIPS Conference 2019 Conference Paper

Graph Neural Tangent Kernel: Fusing Graph Neural Networks with Graph Kernels

• Simon Du
• Kangcheng Hou
• Russ Salakhutdinov
• Barnabas Poczos
• Ruosong Wang
• Keyulu Xu

While graph kernels (GKs) are easy to train and enjoy provable theoretical guarantees, their practical performances are limited by their expressive power, as the kernel function often depends on hand-crafted combinatorial features of graphs. Compared to graph kernels, graph neural networks (GNNs) usually achieve better practical performance, as GNNs use multi-layer architectures and non-linear activation functions to extract high-order information of graphs as features. However, due to the large number of hyper-parameters and the non-convex nature of the training procedure, GNNs are harder to train. Theoretical guarantees of GNNs are also not well-understood. Furthermore, the expressive power of GNNs scales with the number of parameters, and thus it is hard to exploit the full power of GNNs when computing resources are limited. The current paper presents a new class of graph kernels, Graph Neural Tangent Kernels (GNTKs), which correspond to \emph{infinitely wide} multi-layer GNNs trained by gradient descent. GNTKs enjoy the full expressive power of GNNs and inherit advantages of GKs. Theoretically, we show GNTKs provably learn a class of smooth functions on graphs. Empirically, we test GNTKs on graph classification datasets and show they achieve strong performance.

NeurIPS Conference 2019 Conference Paper

On Exact Computation with an Infinitely Wide Neural Net

• Sanjeev Arora
• Simon Du
• Wei Hu
• Zhiyuan Li
• Russ Salakhutdinov
• Ruosong Wang

How well does a classic deep net architecture like AlexNet or VGG19 classify on a standard dataset such as CIFAR-10 when its “width”— namely, number of channels in convolutional layers, and number of nodes in fully-connected internal layers — is allowed to increase to infinity? Such questions have come to the forefront in the quest to theoretically understand deep learning and its mysteries about optimization and generalization. They also connect deep learning to notions such as Gaussian processes and kernels. A recent paper [Jacot et al. , 2018] introduced the Neural Tangent Kernel (NTK) which captures the behavior of fully-connected deep nets in the infinite width limit trained by gradient descent; this object was implicit in some other recent papers. An attraction of such ideas is that a pure kernel-based method is used to capture the power of a fully-trained deep net of infinite width. The current paper gives the first efficient exact algorithm for computing the extension of NTK to convolutional neural nets, which we call Convolutional NTK (CNTK), as well as an efficient GPU implementation of this algorithm. This results in a significant new benchmark for performance of a pure kernel-based method on CIFAR-10, being 10% higher than the methods reported in [Novak et al. , 2019], and only 6% lower than the performance of the corresponding finite deep net architecture (once batch normalization etc. are turned off). Theoretically, we also give the first non-asymptotic proof showing that a fully-trained sufficiently wide net is indeed equivalent to the kernel regression predictor using NTK.

NeurIPS Conference 2019 Conference Paper

Provably Efficient Q-learning with Function Approximation via Distribution Shift Error Checking Oracle

• Simon Du
• Yuping Luo
• Ruosong Wang
• Hanrui Zhang

Q-learning with function approximation is one of the most popular methods in reinforcement learning. Though the idea of using function approximation was proposed at least 60 years ago, even in the simplest setup, i. e, approximating Q-functions with linear functions, it is still an open problem how to design a provably efficient algorithm that learns a near-optimal policy. The key challenges are how to efficiently explore the state space and how to decide when to stop exploring in conjunction with the function approximation scheme. The current paper presents a provably efficient algorithm for Q-learning with linear function approximation. Under certain regularity assumptions, our algorithm, Difference Maximization Q-learning, combined with linear function approximation, returns a near-optimal policy using polynomial number of trajectories. Our algorithm introduces a new notion, the Distribution Shift Error Checking (DSEC) oracle. This oracle tests whether there exists a function in the function class that predicts well on a distribution $\mathcal{D}_1$, but predicts poorly on another distribution $\mathcal{D}_2$, where $\mathcal{D}_1$ and $\mathcal{D}_2$ are distributions over states induced by two different exploration policies. For the linear function class, this oracle is equivalent to solving a top eigenvalue problem. We believe our algorithmic insights, especially the DSEC oracle, are also useful in designing and analyzing reinforcement learning algorithms with general function approximation.

SODA Conference 2019 Conference Paper

Tight Bounds for ℓp Oblivious Subspace Embeddings

• Ruosong Wang
• David P. Woodruff

An ℓ p oblivious subspace embedding is a distribution over r × n matrices n such that for any fixed n × d matrix A, where r is the dimension of the embedding, κ is the distortion of the embedding, and for an n -dimensional vector y, ‖ y ‖ p = (∑ i =1 n | y i |) 1/ p is the ℓ p -norm. Another important property is the sparsity of Π, that is, the maximum number of nonzero entries per column, as this determines the running time of computing Π · A. While for p = 2 there are nearly optimal tradeoffs in terms of the dimension, distortion, and sparsity, for the important case of 1 ≤ p < 2, much less was known. In this paper we obtain nearly optimal tradeoffs for ℓ p oblivious subspace embeddings for every 1 ≤ p < 2. Our main results are as follows: 1. We show for every 1 ≤ p < 2, any oblivious subspace embedding with dimension r has distortion. When r = poly( d ) ≪ n in applications, this gives a κ = Ω( d 1/ p log −2/ p d ) lower bound, and shows the oblivious subspace embedding of Sohler and Woodruff (STOC, 2011) for p = 1 and the oblivious subspace embedding of Meng and Mahoney (STOC, 2013) for 1 < p < 2 are optimal up to poly(log( d )) factors. 2. We give sparse oblivious subspace embeddings for every 1 ≤ p < 2 which are optimal in dimension and distortion, up to poly(log d ) factors. Importantly for p = 1, we achieve r = O ( d log d ), κ = O ( d log d ) and s = O (log d ) non-zero entries per column. The best previous construction with s ≤ poly(log d ) is due to Woodruff and Zhang (COLT, 2013), giving κ = Ω( d 2 poly(log d )) or κ = Ω( d 3/2 · poly(log d )) and r ≥ d · poly(log d ); in contrast our r = O ( d log d ) and κ = O ( d log d ) are optimal up to poly(log( d )) factors even for dense matrices. We also give (1) nearly-optimal ℓ p oblivious subspace embeddings with an expected 1 + ε number of non-zero entries per column for arbitrarily small ε > 0, and (2) the first oblivious subspace embeddings for 1 ≤ p < 2 with O (1)-distortion and dimension independent of n. Oblivious subspace embeddings are crucial for distributed and streaming environments, as well as entrywise ℓ p low rank approximation. Our results give improved algorithms for these applications.

AAAI Conference 2017 Conference Paper

Bounded Rationality of Restricted Turing Machines

• Lijie Chen
• Pingzhong Tang
• Ruosong Wang

Bounded rationality aims to understand the effects of how limited rationality affects decision-making. The traditional models in game theory and multiagent system research, such as finite automata or unrestricted Turing machine, fall short of capturing how intelligent agents make decision in realistic applications. To address this problem, we model bounded rational agents as restricted Turing machines: restrictions on running time and on storage space. We study our model under the context of two-person repeated games. In the case where the running time of Turing machines is restricted, we show that computing the best response of a given strategy is much harder than the strategy itself. In the case where the storage space of the Turing machines is restricted, we show the best response of a space restricted strategy can not be implemented by machines within the same size (up to a constant factor). Finally, we study how these restrictions affect the set of Nash equilibria in infinitely repeated games. We show restricting the agent’s computational resources will give rise to new Nash equilibria.

AAMAS Conference 2017 Conference Paper

Efficient Near-optimal Algorithms for Barter Exchange

• Zhipeng Jia
• Pingzhong Tang
• Ruosong Wang
• Hanrui Zhang

We study polynomial-time clearing algorithms for the barter exchange problem. We put forward a family of carefully designed approximation algorithms with desirable worst-case guarantees. We further apply a series of novel heuristics to implement these algorithms. We demonstrate via kidney exchange data sets that these algorithms achieve near-optimal performances while outperforming the state-of-the-art ILP based algorithms in running time by orders of magnitude.

STOC Conference 2017 Conference Paper

Exponential separations in the energy complexity of leader election

• Yi-Jun Chang
• Tsvi Kopelowitz
• Seth Pettie
• Ruosong Wang
• Wei Zhan

Energy is often the most constrained resource for battery-powered wireless devices and the lion's share of energy is often spent on transceiver usage (sending/receiving packets), not on computation. In this paper we study the energy complexity of Leader Election and Approximate Counting in several models of wireless radio networks. It turns out that energy complexity is very sensitive to whether the devices can generate random bits and their ability to detect collisions . We consider four collision-detection models: Strong-CD (in which transmitters and listeners detect collisions), Sender-CD and Receiver-CD (in which only transmitters or only listeners detect collisions), and No-CD (in which no one detects collisions.)

AAMAS Conference 2017 Conference Paper

K-Memory Strategies in Repeated Games

• Lijie Chen
• Fangzhen Lin
• Pingzhong Tang
• Kangning Wang
• Ruosong Wang
• Shiheng Wang

In this paper, we study k-memory strategies in two-person repeated games. An agent adopting such a strategy makes his decision only based on the action profiles of the previous k rounds. We show that in finitely repeated games, the best response of a k-memory strategy may not even be a constant-memory strategy. However, in infinitely repeated games, one of the best responses against any given k-memory strategy must be k-memory. Our results are enabled by a graph-structural characterization of the best responses of kmemory strategies. We put forward polynomial algorithms to compute best responses.