Pessimistic Minimax Value Iteration finds efficient NE policies from offline data.
problem Finding an approximate Nash equilibrium in offline Markov games with non-uniform coverage.
method Pessimistic Minimax Value Iteration (PMVI) constructs pessimistic value function estimates and solves NEs.
result Established a nearly minimax optimal result for offline Markov games with function approximation.
New method estimates minimizer and minimum value of a regression function.
problem Estimating minimizer and minimum value of a regression function from noisy data.
method Projected gradient descent with gradient estimated by regularized local polynomial algorithm, followed by a rate optimal nonparametric procedure.
result Achieves minimax optimal rates of convergence for smooth and strongly convex functions.
Study on estimating invertible functions with minimax analysis.
problem Minimizing risk of estimating invertible functions on a plane.
method Introduce two types of L 2 L^2 L 2 -risks, derive lower and upper rates for minimax values, develop an asymptotically almost everywhere invertible estimator. result Invertibility does not reduce the complexity of the estimation problem in terms of the rate.
Study minimax-optimal rates for offline decision-making with function approximation.
problem Statistical complexity of offline decision-making with function approximation.
method Near minimax-optimal rates for stochastic contextual bandits and Markov decision processes, using pseudo-dimension and behavior policy.
result Established performance limits and new characterization of behavior policy.
New method for off-policy evaluation in POMDPs using future-dependent value functions.
problem Curse of horizon in off-policy evaluation for POMDPs.
method Develops future-dependent value functions and minimax learning method.
result PAC result and Bellman completeness for the proposed OPE estimator.
New algorithms solve complex minimax problems without needing derivatives.
problem Solving nonconvex-concave minimax problems efficiently.
method Zeroth-order alternating and proximal gradient algorithms.
result Iteration complexity and function value estimation bounds established.
Paper proposes algorithms for solving nonconvex-nonconcave problems with complexity guarantees.
problem Nonconvex-nonconcave minimax problems with PL condition.
method Zeroth-order AGDA and VRAGDA algorithms.
result Iteration complexities for obtaining ε-stationary points.
This paper analyzes how machine learning models resist adversarial attacks in nonparametric regression.
problem Adversarial attacks on machine learning models in nonparametric regression.
method Theoretical analysis of minimax rates of convergence under adversarial sup-norm.
result The minimax rate under adversarial attacks is the sum of two terms: standard rate and deviation of true function.
New method for evaluating policies in complex decision-making models with hidden variables.
problem Evaluating policies in partially observable Markov decision processes with hidden confounders.
method Introduces novel identification methods and minimax estimation techniques for linking target policy's value and observed data distribution.
result Proposes three estimators for off-policy evaluation in POMDPs with latent confounders, demonstrating their effectiveness through nonasymptotic and asymptotic analysis.
We consider the problem of global optimization of an unknown non-convex smooth function with zeroth-order feedback. In this setup, an algorithm is allowed to adaptively query the underlying function at different locations and receives noisy evaluations of function values at the queried points (i.e. the algorithm has ac…
Study confirms learning rates for vector-valued spectral algorithms, proving consistency.
problem Theoretical confirmation of learning rates for vector-valued spectral algorithms.
method Rigorous analysis of learning rates for various vector-valued spectral algorithms, including kernel ridge regression and gradient descent.
result Upper and lower bounds on learning rates for vector-valued spectral algorithms, proving minimax optimality in various scenarios.
New algorithms estimate Q-functions under partial coverage and realizability, improving offline RL guarantees.
problem Offline RL with limited exploration and assumptions about data coverage and Q-function realizability.
method Proposes minimax learning algorithms to estimate soft or vanilla Q-functions with L 2 L^2 L 2 -convergence guarantees. result PAC guarantees for offline RL under partial coverage and realizability conditions.
We study minimax methods for off-policy evaluation (OPE) using value functions and marginalized importance weights. Despite that they hold promises of overcoming the exponential variance in traditional importance sampling, several key problems remain: (1) They require function approximation and are generally biased. Fo…
New method estimates optimal Q-values with better accuracy for specific problems.
problem Estimating optimal Q-values in reinforcement learning is difficult and varies by problem instance.
method Local minimax framework and variance-reduced Q-learning.
result Sharp lower bounds on estimation accuracy for Q-learning.
Papers learn from data to make decisions without interacting, improving on previous methods.
problem Achieving optimal decision-making from offline data with non-linear function approximation.
method Pessimistic Nonlinear Least-Square Value Iteration (PNLSVI) with three innovative components.
result Achieves minimax optimal instance-dependent regret for non-linear function approximation.
The paper optimizes risk-sensitive RL with CVaR, achieving near-minimax-optimal results.
problem Optimizing risk-sensitive reinforcement learning with CVaR objective.
method Developed algorithms for multi-arm bandits and online RL in MDPs, achieving near-minimax-optimal regret.
result Achieved near-minimax-optimal regret of O ( τ − 1 S A K ) O(τ^{-1}\sqrt{SAK}) O ( τ − 1 S A K ) for constant τ τ τ . The paper analyzes kNN density estimation's convergence rates under different conditions.
problem Analyzing convergence rates of kNN density estimation under bounded and unbounded support conditions.
method Examined two cases: bounded support with known and unknown support sets, and unbounded support with smooth density function.
result kNN density estimation is minimax optimal under certain conditions and better than kernel density estimation in some cases.
MOPI optimizes flexible set-valued mappings to achieve superior shape adaptivity in conformal prediction.
problem Challenges in achieving valid conditional coverage in conformal prediction.
method Minimax Optimization Predictive Inference (MOPI) framework that optimizes over a flexible class of set-valued mappings.
result MOPI achieves superior shape adaptivity and maintains a principled connection to mean squared coverage error.
Study minimax regret in bilateral trade with heavy-tailed valuations.
problem Minimizing regret in bilateral trade with infinite variance valuations.
method Extended self-bounding property, truncated-mean estimation, epoch-based algorithm.
result Achieves regret bound of O ( T 1 − 2 β ( p − 1 ) / ( β p + d ( p − 1 ) ) ) O(T^{1-2β(p-1)/(βp + d(p-1))}) O ( T 1 − 2 β ( p − 1 ) / ( β p + d ( p − 1 )) ) under specific conditions. New algorithm FLUTE achieves uniform-PAC convergence in RL with linear approx.
problem RL with linear function approximation lacks uniform-PAC guarantees.
method FLUTE algorithm with minimax value function estimator and multi-level partition scheme.
result Uniform-PAC convergence to optimal policy with high probability.
New RL algorithm achieves nearly optimal performance for linear MDPs.
problem Optimal reinforcement learning for episodic linear MDPs.
method Weighted linear regression with variance estimator and rare-switching policy.
result Achieves nearly minimax optimal regret i l d e O ( d H 3 K ) ilde O(d\sqrt{H^3K}) i l d e O ( d H 3 K ) . New algorithm achieves optimal regret in average reward MDPs without prior bias information.
problem Achieving optimal regret in average reward MDPs with computational efficiency and without prior bias information.
method Projective Mitigated Extended Value Iteration (PMEVI) to compute bias-constrained optimal policies efficiently.
result First tractable algorithm with minimax optimal regret of O ~ ( s p ( h ∗ ) S A T ) \widetilde{\mathrm{O}}(\sqrt{\mathrm{sp}(h^*) S A T}) O ( sp ( h ∗ ) S A T ) . We study the problem of alleviating the instability issue in the GAN training procedure via new architecture design. The discrepancy between the minimax and maximin objective values could serve as a proxy for the difficulties that the alternating gradient descent encounters in the optimization of GANs. In this work, we…
New algorithms tackle robust RL with linear models, revealing unique challenges.
problem Distributionally robust offline RL with uncertainty in dynamics.
method Proposes minimax optimal and computationally efficient algorithms using novel function approximation mechanisms.
result Function approximation in robust offline RL is distinct and harder than in standard offline RL.
Paper tackles linear models with missing values, achieving minimax optimal results.
problem Missing values in real-world data complicate linear model learning.
method Proposes a rigorous setting and a new algorithm leveraging missing data distribution.
result Derives minimax optimal adaptive risk bounds for predictions with missing values.
New RL method nearly optimally learns policies with generative models.
problem Finding optimal policies in reinforcement learning with generative models.
method Mirror descent value iteration with KL divergence and entropy regularization.
result The method is nearly minimax-optimal for small ε \varepsilon ε -optimal policies. Study on Q Q Q -function estimation for continuous state-action MDPs, deriving rates and conditions.
problem Estimating Q Q Q -function in off-policy evaluation for continuous state-action Markov decision processes. method Reformulated as nonparametric instrumental variables (NPIV) problem, derived minimax lower bounds, proposed sieve two-stage least squares estimator.
result First minimax lower bounds for Q Q Q -function and its derivatives in sup-norm and L 2 L^2 L 2 -norm, same as classical nonparametric regression. Paper analyzes error bounds for learning with vector-valued RF, improving existing analyses.
problem Learning with vector-valued random features in infinite-dimensional settings.
method Direct analysis of risk functional, avoiding random matrix theory.
result Strong consistency and minimax optimal convergence rates established.
Estimation of functions of d d d variables is considered using ridge combinations of the form ∑ k = 1 m c 1 , k φ ( ∑ j = 1 d c 0 , j , k x j − b k ) \textstyle\sum_{k=1}^m c_{1,k} φ(\textstyle\sum_{j=1}^d c_{0,j,k}x_j-b_k) ∑ k = 1 m c 1 , k φ ( ∑ j = 1 d c 0 , j , k x j − b k ) where the activation function φ φ φ is a function with bounded value and derivative. These include single-hidden layer neural networks, polynomials, …
New algorithm reduces contextual bandits to efficient regression.
problem Developing efficient algorithms for contextual bandits with general function classes.
method Reduction from contextual bandits to online regression with oracle.
result First universal and optimal reduction with no overhead.
Minimax linkage was first introduced by Ao et al. [3] in 2004, as an alternative to standard linkage methods used in hierarchical clustering. Minimax linkage relies on distances to a prototype for each cluster; this prototype can be thought of as a representative object in the cluster, hence improving the interpretabil…
Study online learning with set-valued feedback, showing differences between deterministic and randomized approaches.
problem Online learning with set-valued feedback, where labels are sets rather than single labels.
method Introduced new combinatorial dimensions (Set Littlestone and Measure Shattering) to characterize learnability.
result Characterized deterministic and randomized online learnability, and established bounds for various learning settings.
Improves RL generalization by minimizing adversarial risk.
problem Overfitting to training environments and poor generalization to unseen scenarios.
method Introduces minimax formulation and distributional framework to RL.
result Trained policy shows improved generalization to different environments.
This paper optimizes off-policy evaluation in reinforcement learning with function approximation.
problem Estimating cumulative value of a new policy from logged data generated by an unknown policy.
method Regression-based fitted Q iteration method, equivalent to estimating conditional mean embedding of transition operator.
result The method is minimax-optimal, with nearly minimal estimation error.
This paper studies continuum-armed bandits under Besov smoothness conditions and derives minimax rates.
problem Optimizing an unknown function with limited evaluations.
method Studies continuum-armed bandits under Besov smoothness conditions and derives minimax rates.
result Minimax rates over Besov spaces are identical to those over the smallest Hölder space into which Besov spaces embed.
We demonstrate an equivalence between reproducing kernel Hilbert space (RKHS) embeddings of conditional distributions and vector-valued regressors. This connection introduces a natural regularized loss function which the RKHS embeddings minimise, providing an intuitive understanding of the embeddings and a justificatio…
Paper tackles gradient-free minimax optimization with variance reduction for faster convergence.
problem Gradient-free minimax optimization problems in machine learning.
method Variance reduction technique to design a novel zeroth-order gradient descent ascent algorithm.
result Achieves the best known query complexity of O(κ(d₁ + d₂)ε⁻³), outperforming previous methods.
We consider the learning algorithms under general source condition with the polynomial decay of the eigenvalues of the integral operator in vector-valued function setting. We discuss the upper convergence rates of Tikhonov regularizer under general source condition corresponding to increasing monotone index function. T…
New RL algorithm tackles nonstationary MDPs with linear approximations and varying rewards.
problem Nonstationary reinforcement learning with evolving reward and state transition functions.
method Developed a new algorithm LSVI-UCB-Restart with periodic restart, and parameter-free Ada-LSVI-UCB-Restart for unknown variation budgets.
result First minimax dynamic regret lower bound for nonstationary linear MDPs and linear MDPs lower bound.
Develops new instance-optimality concepts in differential privacy.
problem Improving privacy guarantees in statistical estimation.
method Introduces local minimax risk and unbiased mechanisms, and develops inverse sensitivity mechanisms.
result Inverse sensitivity mechanisms are nearly instance optimal for a wide range of functions.
New algorithm reduces RL complexity with low switching costs.
problem Exploration-exploitation dilemma in RL with complex models.
method Monotonic Q-Learning with Upper Confidence Bound (MQL-UCB) for RL with general function approximation.
result Achieves minimax optimal regret of O ( d H K ) O(d\sqrt{HK}) O ( d H K ) and near-optimal policy switching cost. We consider random-design linear prediction and related questions on the lower tail of random matrices. It is known that, under boundedness constraints, the minimax risk is of order d / n d/n d / n in dimension d d d with n n n samples. Here, we study the minimax expected excess risk over the full linear class, depending on the dist…
New methods estimate transport-growth pairs in unbalanced optimal transport.
problem Statistical guarantees for Monge-type estimation in unbalanced optimal transport remain limited.
method Developed two estimators for transport-growth pairs under different setups.
result Achieved minimax optimal rate for estimation of transport-growth pairs.
Neural networks minimize error with shallow ReLU models for function estimation.
problem Estimating unknown functions from noisy data.
method Minimizing squared errors plus weight decay regularization.
result Neural network estimators are minimax optimal up to logarithmic factors.
Despite the great empirical success of deep reinforcement learning, its theoretical foundation is less well understood. In this work, we make the first attempt to theoretically understand the deep Q-network (DQN) algorithm (Mnih et al., 2015) from both algorithmic and statistical perspectives. In specific, we focus on …
Study shows how to effectively predict functions on manifolds using kernel methods.
problem Regression on manifolds with limited data.
method Reproducing kernel Hilbert space methods, Weyl law, effective dimension.
result Kernel regression estimator yields minimax-optimal error bounds controlled by effective dimension.
The paper analyzes the efficiency of gradient estimation methods in noisy function evaluations.
problem Estimating gradients of smooth functions using noisy function evaluations.
method Information-theoretic lower bounds and finite difference method analysis.
result The finite difference method is not minimax optimal, suggesting room for improvement in gradient estimation.
Estimates BV functions from noisy data using Voronoi diagrams.
problem Estimating multivariate BV functions from scattered noisy data.
method Form Voronoi diagram, solve optimization problem with discrete TV regularization.
result Voronoigram is minimax rate optimal for BV functions.