Researchers estimate optimal PAC-Bayes bounds using Hamiltonian Monte Carlo.
problem Estimating tight PAC-Bayes bounds with restricted posterior families.
method Sampling from optimal Gibbs posterior using Hamiltonian Monte Carlo, estimating KL divergence, and proposing high-probability bounds.
result Significant tightness gaps in PAC-Bayes bounds, up to 5-6% in some cases.
PAC-Bayesian theory applied to learning optimization algorithms with generalization guarantees.
problem Learning optimization algorithms with provable generalization guarantees and explicit trade-offs.
method PAC-Bayes theory applied to learning-to-optimize, reformulating the learning procedure into a one-dimensional minimization problem.
result Learned optimization algorithms outperform deterministic worst-case analysis algorithms, even in the limit case of guaranteed convergence.
PACOH improves meta-learning with theoretical guarantees and practical efficiency.
problem Meta-learning's generalization to unseen tasks is poorly understood, especially with limited meta-training tasks.
method PAC-Bayesian framework for deriving generalization bounds and developing PAC-optimal meta-learning algorithms.
result PACOH yields state-of-the-art performance in predictive accuracy and uncertainty estimation.
Paper refines PAC-Bayes bounds for bandit problems.
problem Improving probabilistic bounds for off-policy learning.
method Optimizes PAC-Bayesian bounds using a new parameter optimization approach.
result Provides two parameter-free PAC-Bayes bounds that nearly match optimal rates.
New algorithm optimizes PAC-Bayes bound without surrogate loss.
problem Mismatch between optimisation objective and generalisation bound in stochastic neural networks.
method Proposes a novel training algorithm that optimizes the PAC-Bayesian bound directly.
result Empirical results show improved performance over existing PAC-Bayesian training methods.
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.
Statistical performance bounds for reinforcement learning (RL) algorithms can be critical for high-stakes applications like healthcare. This paper introduces a new framework for theoretically measuring the performance of such algorithms called Uniform-PAC, which is a strengthening of the classical Probably Approximatel…
New PAC-Bayes training method improves model generalization for unbounded loss.
problem Improving generalization of complex models under unbounded loss.
method Established new PAC-Bayes bound for unbounded loss, jointly training prior and posterior.
result Outperforms existing PAC-Bayes training algorithms and matches ERM accuracy.
PAC-Bayes bound requires prior to place mass on high-performing predictors.
problem Explaining generalization in machine learning.
method Analyzing necessary conditions for PAC-Bayes bounds to provide meaningful generalization guarantees.
result Achieving a target generalisation level requires the prior to place sufficient mass on high-performing predictors.
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 reduce communication costs in collaborative learning.
problem Reducing communication costs in collaborative learning.
method Distributed boosting and adaptation to classification noise.
result Communication-efficient algorithms for collaborative PAC learning robust to noise.
New algorithm reduces contextual bandit identification to argmax calls.
problem Best-arm identification in stochastic contextual bandits.
method Instance-optimal PAC algorithm using argmax oracle calls.
result First instance-dependent PAC sample complexity for contextual bandits.
New PAC-Bayes bounds for unbounded losses using Cramér-Chernoff techniques.
problem Developing bounds for unbounded losses in PAC-Bayesian settings.
method Introducing a new PAC-Bayes oracle bound using Cramér-Chernoff bounds and controlling random variable tails.
result Our bounds generalize and improve upon previous results, providing more informative and potentially tighter bounds.
This work establishes a new upper bound on the number of samples sufficient for PAC learning in the realizable case. The bound matches known lower bounds up to numerical constant factors. This solves a long-standing open problem on the sample complexity of PAC learning. The technique and analysis build on a recent brea…
PAC-Bayes guarantees for heavy-tailed losses with robust risk estimator.
problem Learning with heavy-tailed losses.
method Derive PAC-Bayesian learning guarantees using heavy-tailed losses and a novel optimal Gibbs posterior.
result Achieves nearly sub-Gaussian statistical error with robust risk estimator.
Deep neural networks are optimal for dependent data using PAC-Bayes bounds.
problem Optimizing deep neural networks for dependent data.
method PAC-Bayes oracle inequalities and Bernstein inequality.
result Upper and lower bounds match, proving minimax optimality.
New PAC-Bayes bounds use Wasserstein distances to improve generalization.
problem Lack of geometric properties in existing PAC-Bayes bounds.
method Developed new PAC-Bayes bounds with Wasserstein distances.
result Optimization guarantees translate to good generalization abilities.
Develops a new PAC-Bayesian bound for evaluating randomness in predictors.
problem Lack of a PAC-Bayesian bound for deterministic predictors and continuous loss functions.
method PAC-Bayesian transportation bound unifying PAC-Bayesian and chaining methods.
result First PAC-Bayesian bound relating risks of any two predictors by their distance.
New PAC-Bayesian bounds for multi-view learning using Rényi divergence.
problem Applying PAC-Bayesian theory to multi-view learning.
method Introducing novel PAC-Bayesian bounds based on Rényi divergence for multi-view learning.
result Efficient optimization algorithms that align with theoretical bounds.
Majority-of-Three is Optimal
problem Optimality of Voting Learners
method Majority vote of three classifiers
result Proves optimality for the simplest voting scheme
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.
New algorithms minimize PAC-Bayesian C-Bound for majority voting, leading to scalable and accurate predictors.
problem Improving majority vote classifiers using PAC-Bayesian bounds.
method Directly optimizing PAC-Bayesian guarantees on the C-Bound with gradient descent.
result Self-bounding majority vote learning algorithms with scalable and accurate predictors.
PAC-Bayesian framework for fairness in stochastic and deterministic classifiers.
problem Theoretical guarantees on fairness for balancing predictive risk and fairness constraints.
method PAC-Bayesian framework for both stochastic and deterministic classifiers, covering a broad class of fairness measures.
result Derives generalization bounds for fairness, demonstrating tightness with empirical evaluation.
Paper proposes efficient algorithms for bandit problems with costly sampling.
problem Maximizing expectation function over a finite set with high sampling cost.
method Proposes naive and adaptive stochastic bandit algorithms for PAC solution.
result Adaptive algorithm outperforms naive in terms of number of samples.
Improved PAC guarantees for multi-agent reinforcement learning with noisy communication.
problem Improving exploration in cooperative multi-agent reinforcement learning with communication constraints.
method Develops PAC guarantees for multiple concurrent MDPs with noisy and resource-limited communication.
result Theoretical and empirical improvements in sample complexity for information fusion.
Bayesian framework reduces online optimization regret.
problem Sequential optimization in dynamic environments with bounded losses.
method PAC-Bayes theory and Bayesian updating principles.
result Achieves O ( T ) \mathcal{O}(\sqrt{T}) O ( T ) regret for bounded losses. Characterizes statistical complexity of realizable regression in PAC and online learning.
problem Understanding the statistical complexity of realizable regression in both PAC and online learning settings.
method Introduces minimax instance optimal learners, novel and combinatorial dimensions to characterize learnability.
result Characterizes which classes of real-valued predictors are learnable and provides necessary conditions for learnability.
Paper improves PAC-Bayes bounds for various loss types.
problem Improving PAC-Bayes bounds for different types of losses.
method Introducing new high-probability PAC-Bayes bounds for bounded and general tail behaviors losses, and extending to anytime-valid bounds.
result New fast-rate and mixed-rate bounds for losses with bounded ranges, and parameter-free bounds for losses with general tail behaviors.
New bounds for model generalization under deterministic gradient descent.
problem Establishing generalization bounds for models trained with gradient descent methods.
method PAC-Bayesian bounds for deterministic optimisation algorithms.
result Fully computable bounds that depend on initial distribution and Hessian.
The paper improves risk certificate tightness for neural networks using PAC-Bayes bounds.
problem Improving the usability of risk certificates for neural networks based on PAC-Bayes bounds.
method Theoretical contributions including KL divergence bounds, efficient methodology for optimization, and methods for optimizing non-differentiable objectives.
result First non-vacuous generalization bounds on CIFAR-10 for neural networks.
New PAC-Bayesian bounds improve Sliced-Wasserstein distances.
problem Improving statistical properties of Sliced-Wasserstein distances.
method Leveraging PAC-Bayesian theory to provide bounds and learning procedures.
result PAC-Bayesian generalization bounds for adaptive SW distances.
Bayes-CPACE optimally explores continuous BAMDPs.
problem Model uncertainty in continuous state and action spaces.
method Covering state-belief-action space with samples, exploiting Lipschitz continuity.
result Near-optimal value function computed efficiently.
Improved PAC learning algorithm for multiclass classification with bandit feedback.
problem Efficiently learning multiclass classification with limited feedback.
method Novel algorithm using stochastic optimization and Frank-Wolfe updates.
result Improved sample complexity bounds for multiclass PAC learning.
Study compares chi-squared divergence and KL-divergence posteriors for PAC-Bayesian bounds.
problem Investigates optimal posteriors for PAC-Bayesian bounds using chi-squared divergence.
method Analyzes bounds for three distance functions, derives FP equations for computation.
result Chi-squared divergence based posteriors have weaker bounds and worse test errors.
Paper develops PAC-Bayes bounds for unknown linear systems.
problem Learning controllers for unknown stochastic linear discrete-time systems.
method PAC-Bayes framework for data-dependent high probability bounds.
result Proposes efficient learning algorithms with theoretical guarantees.
PAC-Bayesian matrix completion with a spectral scaled Student prior offers efficient inference.
problem Matrix completion with underlying low-rank structure.
method Spectral scaled Student prior and PAC-Bayesian bounds.
result Minimax-optimal oracle inequality for model misspecification and general sampling distribution.
A new hyperparameter optimization method reduces overfitting.
problem Overfitting in hyperparameter optimization.
method PAC-Bayes bound minimization using gradient-based algorithm.
result Significant reduction in out-of-sample error.
We show that the class of strongly connected graphical models with treewidth at most k can be properly efficiently PAC-learnt with respect to the Kullback-Leibler Divergence. Previous approaches to this problem, such as those of Chow ([1]), and Ho gen ([7]) have shown that this class is PAC-learnable by reducing it to …
Optimizes pruning masks for neural networks using probabilistic fine-tuning and PAC-Bayes bounds.
problem Improving neural network performance through adaptive pruning of weights.
method Optimizes stochastic pruning masks by minimizing expected loss, considering data-adaptive regularization and feature alignment.
result Probabilistic fine-tuning leads to improved test error over baseline methods in neural networks.
Develops a theory to make learning solutions fair and safe.
problem Ensuring learning solutions are unbiased and safe in critical applications.
method Generates a generalization theory based on PAC learning framework, introduces constrained learning algorithm.
result Proves that constrained learning is as learnable as unconstrained learning, provides practical algorithm.
New method improves meta-learning scalability and generalization.
problem Overfitting in meta-learning with limited related tasks.
method PAC-Bayesian theory and optimal hyper-posterior (PACOH).
result PACOH provides better performance guarantees and scalability.
New bound improves on weighted majority vote risk estimation.
problem Improving risk estimation for weighted majority vote.
method Novel Chebyshev-Cantelli inequality and PAC-Bayes-Bennett inequality.
result New bounds improve on existing methods.
Study shows mean-field approximation fails to improve PAC-Bayes bounds for neural networks.
problem Understanding why overparametrized neural networks achieve low risk and zero empirical risk.
method Optimized PAC-Bayes bounds using variational inference (VI), investigating mean-field approximation.
result Mean-field approximation does not provide significant improvements in PAC-Bayes bounds for neural networks.
Optimal sample complexity for learning Plackett-Luce models.
problem PAC-learning good items from subsetwise feedback in Plackett-Luce models.
method Algorithm based on a wrapper around a PAC winner-finding algorithm, adapting to instance hardness.
result Optimal instance-dependent sample complexity for best arm identification.
We show that Entropy-SGD (Chaudhari et al., 2017), when viewed as a learning algorithm, optimizes a PAC-Bayes bound on the risk of a Gibbs (posterior) classifier, i.e., a randomized classifier obtained by a risk-sensitive perturbation of the weights of a learned classifier. Entropy-SGD works by optimizing the bound's p…
New algorithm reduces worst-case sample complexity for learning best arm.
problem Identifying a best arm with confidence in multi-armed bandit settings.
method Proposed a new ( ε , δ ) (ε,δ) ( ε , δ ) -PAC learning algorithm for multi-armed bandits. result Algorithm achieves optimal sample complexity for ( ε , δ ) (ε,δ) ( ε , δ ) -learning. New reinforcement learning algorithm achieves instance-optimal sample complexity.
problem Achieving low regret and identifying optimal policies in reinforcement learning.
method A novel planning-based algorithm that explicitly accounts for state visitation distributions.
result The proposed algorithm attains nearly minimax optimal sample complexity, improving over worst-case bounds.
Study risk-sensitive reinforcement learning with optimized certainty equivalents.
problem Risk-sensitive reinforcement learning in finite discounted MDPs.
method Analyzed a simple model-based approach and derived PAC sample complexity bounds.
result Established tight sample complexity bounds for value and policy learning.