Algorithm reduces regret in misspecified linear contextual bandits.
problem Misspecified linear contextual bandits with bounded misspecification.
method Data selection scheme for online regression, leveraging uncertainty.
result Regret bound of O ~ ( d 2 / Δ ) \tilde O(d^2/Δ) O ~ ( d 2 /Δ ) when ζ ≤ O ~ ( Δ / d ) ζ \leq \tilde O(Δ/\sqrt{d}) ζ ≤ O ~ ( Δ/ d ) . Logarithmic regret achieved in Q-learning with positive gap.
problem Achieving logarithmic cumulative regret in Q-learning with positive sub-optimality gap.
method Optimistic Q-learning with logarithmic regret bound.
result Logarithmic cumulative regret bound proven for optimistic Q-learning.
Unified hybrid RL algorithm improves online RL performance with offline data.
problem Improving reinforcement learning performance with limited online data.
method A unified hybrid RL algorithm combining offline and online data.
result Unified algorithm achieves state-of-the-art results in sub-optimality gap and online learning regret.
New algorithm reduces sample complexity for planning in MDPs.
problem Planning in MDPs with unknown transitions.
method MDP-GapE, a trajectory-based MCTS algorithm.
result Proves upper bound on sample complexity in terms of sub-optimality gaps.
Paper tackles sample-efficient RL for linearly realizable MDPs with limited revisiting.
problem Sample-efficient reinforcement learning for linearly realizable MDPs with limited revisiting.
method Develops a new sampling protocol that allows for backtracking and revisiting states in a controlled manner.
result Achieves polynomial sample complexity scaling with feature dimension, horizon, and inverse sub-optimality gap.
We study the performance of the certainty equivalent controller on Linear Quadratic (LQ) control problems with unknown transition dynamics. We show that for both the fully and partially observed settings, the sub-optimality gap between the cost incurred by playing the certainty equivalent controller on the true system …
New measure quantifies function similarity for optimization.
problem Measuring similarity between functions for optimization.
method Quantifies sub-optimality gaps and operation rules.
result Unified measure for various functional similarities.
Logarithmic regret achieved in RL with linear function approximation.
problem Achieving logarithmic regret in reinforcement learning with linear function approximation.
method LSVI-UCB for linear MDP assumption, UCRL-VTR for linear mixture MDP assumption.
result Logarithmic regret bounds established for RL with linear function approximation.
Study gap-dependent regret bounds for risk-sensitive RL.
problem Risk-sensitive reinforcement learning with entropic risk measure.
method Propose cascaded gaps to adapt to problem structures, derive regret bounds.
result Exponential improvement over existing bounds in appropriate settings.
Optimal algorithms identify non-dominated arms in multi-output linear bandit models.
problem Identifying the Pareto Set in multi-output linear bandit models.
method Design-based algorithms for Pareto Set Identification (PSI) in a structured multi-output linear bandit model.
result Nearly optimal guarantees in both fixed-budget and fixed-confidence settings.
AdMRL improves meta-reinforcement learning by minimizing worst-case sub-optimality gap.
problem Meta-reinforcement learning's sensitivity to task distribution shift.
method Model-based adversarial approach with minimax objective and alternating optimization.
result Efficacy in worst-case performance, generalization to out-of-distribution tasks, and sample efficiency.
We consider the dynamic assortment optimization problem under the multinomial logit model (MNL) with unknown utility parameters. The main question investigated in this paper is model mis-specification under the ε \varepsilon ε -contamination model, which is a fundamental model in robust statistics and machine learning. In…
This paper tackles robust policy learning under concept drifts, improving upon existing methods.
problem Tackles robust policy learning under concept drifts, improving upon existing methods.
method Develops a doubly-robust estimator and a learning algorithm to maximize policy value within a given policy class.
result The proposed algorithm achieves sub-optimality gap of the order κ ( Π ) n − 1 / 2 κ(Π)n^{-1/2} κ ( Π ) n − 1/2 , demonstrating substantial improvement over existing benchmarks. New model predicts optimal control for restless bandit problems.
problem Optimizing control in restless bandit problems with minimal assumptions.
method Model Predictive Control with rolling horizon linear programming.
result Sub-optimality gap of O(1/√N) under general conditions, and exp(-Ω(N)) under local-stability condition.
In this paper, we propose to combine imitation and reinforcement learning via the idea of reward shaping using an oracle. We study the effectiveness of the near-optimal cost-to-go oracle on the planning horizon and demonstrate that the cost-to-go oracle shortens the learner's planning horizon as function of its accurac…
We consider the problem of \textit{best arm identification} with a \textit{fixed budget T T T }, in the K K K -armed stochastic bandit setting, with arms distribution defined on [ 0 , 1 ] [0,1] [ 0 , 1 ] . We prove that any bandit strategy, for at least one bandit problem characterized by a complexity H H H , will misidentify the best arm with pr…
Previous work on recommender systems mainly focus on fitting the ratings provided by users. However, the response patterns, i.e., some items are rated while others not, are generally ignored. We argue that failing to observe such response patterns can lead to biased parameter estimation and sub-optimal model performanc…
Improved reinforcement learning with emergency stops.
problem Reducing exploration in reinforcement learning.
method Emergency stop mechanisms to reduce sample complexity.
result Significant improvement in sample complexity and speed.
Paper bounds PAC RL sample complexity in deterministic MDPs.
problem Identify ε-optimal policy with high probability.
method Proposes nearly matching upper and lower bounds on sample complexity, introduces deterministic return gap, uses graph-theoretical concepts and maximum-coverage exploration.
result First nearly matching upper and lower bounds on sample complexity for PAC RL in deterministic MDPs.
New algorithms identify best policies in discounted linear MDPs efficiently.
problem Identifying the best policy in discounted linear MDPs with limited samples.
method Derive lower bounds and devise simple yet near-optimal algorithms.
result Upper bound on sample complexity matches existing bounds.
Optimizes ranking of top-k players from partial comparison data.
problem Identifying the top-k players from incomplete pairwise comparisons.
method Maximum Likelihood Estimator (MLE) and Spectral Method.
result MLE achieves optimal partial and exact recovery, while Spectral Method is sub-optimal.
New algorithm for identifying optimal arms in stochastic bandit problems.
problem Optimal arm identification in stochastic bandit problems with many arms.
method Characterized optimal learning rates and provided algorithms with matching bounds.
result Lower bounds and matching upper bounds for cumulative regret and best-arm identification.
We consider effort allocation in crowdsourcing, where we wish to assign labeling tasks to imperfect homogeneous crowd workers to maximize overall accuracy in a continuous-time Bayesian setting, subject to budget and time constraints. The Bayes-optimal policy for this problem is the solution to a partially observable Ma…
This work develops a unified framework for RLHF with general f f f -divergence regularization.
problem Theoretical understanding of general f f f -divergence regularization in RLHF. method Holistic approach across f f f -divergence class, two algorithms based on distinct sampling principles. result Provably efficient algorithms with O ( log T ) O(\log T) O ( log T ) regret and O ( 1 / T ) O(1/T) O ( 1/ T ) sub-optimality gap. The paper tackles system identification via Hankel nuclear norm regularization, improving estimation rates and singular value gaps.
problem Identifying low-order linear systems from limited data.
method Hankel nuclear norm regularization to encourage low-rankness of the Hankel matrix.
result Hankel regularization enables optimal system recovery with fewer observations and better estimation rates.
Algorithm finds a simplified model for reinforcement learning under agent limitations.
problem Finding a simple model that approximates the true model for reinforcement learning.
method Uses rate-distortion theory to compute an approximately-value-equivalent, lossy compression of the environment.
result Proves an information-theoretic, Bayesian regret bound for the algorithm.
Approximate models help RL by reducing policy search space.
problem How much does an approximate model help in learning near-optimal policies in RL?
method Study sample complexity in RL with an approximate model, providing an algorithm and a lower bound.
result An approximate model can reduce sample complexity by eliminating sub-optimal actions.
IRM fails to capture natural invariances on simple problems.
problem IRM fails to capture natural invariances on simple problems.
method IRM formulation and practical linear variant.
result IRM can lead to worse generalization than unconstrained ERM.
Aims to create safe reinforcement learning policies by considering individual harm.
problem Optimal policies for a population may harm certain individuals.
method Formalizes individual harm, proposes a two-stage procedure, and establishes finite-sample properties.
result Learned policies maximize expected return while minimizing harm.
We consider a finite-armed structured bandit problem in which mean rewards of different arms are known functions of a common hidden parameter θ ∗ θ^* θ ∗ . Since we do not place any restrictions of these functions, the problem setting subsumes several previously studied frameworks that assume linear or invertible reward funct…
SAIL learns from sub-optimal demonstrations to improve sample efficiency in sparse reward tasks.
problem Reducing sample complexity in sparse-rewarded tasks.
method Self-Adaptive Imitation Learning (SAIL) that exploits sub-optimal demonstrations and efficient exploration.
result Significantly improved sample efficiency and better final performance across various tasks.
The paper sets fundamental limits for ERM in high dimensions.
problem Understanding statistical accuracy of ERM in high-dimensional settings.
method Sharp performance characterizations and tight lower bounds derived for generalized linear models.
result Optimal tuning of loss function and regularization parameter.
Wide networks are often believed to have a nice optimization landscape, but what rigorous results can we prove? To understand the benefit of width, it is important to identify the difference between wide and narrow networks. In this work, we prove that from narrow to wide networks, there is a phase transition from havi…
Improves robust transfer learning with side information.
problem Addressing environmental shift in MDPs with side information.
method Estimate-centered uncertainty sets with side information integration.
result Improved robust policy with reduced sub-optimality gap.
New proof shows deep neural nets can have sub-optimal local minima.
problem Can over-parameterization eliminate sub-optimal local minima in deep neural networks?
method Counter-example with generic input data and non-linear activation functions.
result Sub-optimal local minima exist in deep neural networks regardless of width.
LinMED is a new linear bandit algorithm with near-optimal regret bound.
problem Optimizing decision-making in linear bandit problems with sub-Gaussian distributions.
method LinMED is a randomized linear bandit algorithm with closed-form arm sampling probabilities.
result LinMED achieves a near-optimal regret bound of d n d\sqrt{n} d n up to logarithmic factors. Two efficient algorithms improve online item recommendation for large user-item matrices.
problem Low rank matrix completion in an online setting with noisy rewards.
method Phased algorithms extsc{PhasedClusterElim} and extsc{DeterminantElim} for item recommendation.
result Near-optimal per-user regret guarantees for item recommendation.
Paper establishes first instance-dependent lower bound for PAC reinforcement learning.
problem Identifying near-optimal policies in tabular MDPs with minimal samples.
method Proposes instance-dependent lower bound for sample complexity.
result Lower bound closely matches PEDEL algorithm's sample complexity.
LEARN-SAM improves RL from sub-optimal demonstrations by localizing expert policies and selectively using demonstrations.
problem Improving RL from sub-optimal or sparse demonstrations.
method Local Ensemble and Reparameterization with Split and Merge of expert policies (LEARN-SAM).
result LEARN-SAM boosts learning speed and accuracy by selectively using demonstrations.
LoCo-RLHF models diverse human feedback with contextual information.
problem Heterogeneous human feedback from diverse contexts and preferences.
method Low-rank contextual preference model, PRS policy.
result LoCo-RLHF achieves tighter sub-optimality gap than existing methods.
ELM improves neural model embeddings for long-tail learning.
problem Learning skewed label distributions in neural models.
method Enforces margins in logit space and regularizes embedding distribution.
result ELM reduces generalization gap and tightens tail class embeddings.
The paper compares theoretical and empirical performance of imputation methods for missing data.
problem Missing data in real-world datasets.
method Contrast of theoretical and empirical imputation methods for prediction.
result Mean-imputation is asymptotically optimal for prediction, while mode-imputation is sub-optimal.
We address reinforcement learning problems with finite state and action spaces where the underlying MDP has some known structure that could be potentially exploited to minimize the exploration rates of suboptimal (state, action) pairs. For any arbitrary structure, we derive problem-specific regret lower bounds satisfie…
Study minimax off-policy evaluation in multi-armed bandits with known and unknown behavior policies.
problem Evaluate policies in multi-armed bandits with unknown behavior policies.
method Develop minimax rate-optimal procedures for known and unknown behavior policies, including the Switch estimator and Chebyshev polynomial-based estimator.
result Plug-in estimator achieves optimal competitive ratio up to a logarithmic factor when behavior policy is unknown.
Neural networks outperform NTK on compositional tasks, revealing a complexity gap.
problem Understanding the performance gap between neural networks and NTK on tasks with compositional structure.
method Characterized Fourier and architectural complexities, and analyzed the minimax rates of the architecture class.
result The NTK estimator is exponentially sub-optimal compared to the minimax floor when complexities decouple.
In autonomous vehicle (AV) control, allowing mistakes can be quite dangerous and costly in the real world. For this reason we investigate methods of training an AV without allowing the agent to explore and instead having a human explorer collect the data. Supervised learning has been explored for AV control, but it enc…
Flexible algorithms for maximizing rewards in structured bandits.
problem Reward maximization in structured stochastic multi-armed bandit problems.
method Asymptotically optimal algorithms using iterative saddle-point solvers.
result Achieves optimal performance with minimal computational burden.
Paper improves risk bound for MTL with graph-dependent data.
problem Sub-optimal risk bound in multi-task learning with graph-dependent data.
method Proposes a new Bennett-type inequality and develops new Talagrand-type inequality and local fractional Rademacher complexity.
result Derives a sharper risk bound of O ( log n n ) O(\frac{\log n}{n}) O ( n l o g n ) .