CJE calibrates cheap LLM judges against an oracle, achieving high accuracy at a fraction of the cost.
problem Inexpensive LLM judges can produce biased rankings, leading to unreliable outcomes.
method CJE uses a small oracle to calibrate cheap scores, then evaluates at scale with valid uncertainty.
result CJE achieves 99% pairwise ranking accuracy at 14x lower cost compared to a 16x oracle/judge cost ratio.
Unified approach for quantum and classical learning from evaluation oracles.
problem Learning from evaluation oracles in quantum and classical settings.
method Inspired by Kearns' SQ and Valiant's weak evaluation oracle, a unified framework is established.
result Characterizes query complexity for learning linear function classes and extends learnability results for quantum circuits.
A new Bayesian method optimizes time-dependent expensive functions with lookahead.
problem Maximizing a time-dependent, expensive oracle with limited evaluations.
method Recursive, two-step lookahead expected payoff (r2LEY) acquisition function.
result r2LEY outperforms myopic methods in synthetic and real-world datasets.
Bayesian-guided method selects optimal design from large candidate pool.
problem Optimizing complex structures with high-fidelity evaluations.
method Bayesian active learning with surrogate modeling.
result Optimal design identified with minimal oracle evaluations.
We develop theory for using heuristics to solve computationally hard problems in differential privacy. Heuristic approaches have enjoyed tremendous success in machine learning, for which performance can be empirically evaluated. However, privacy guarantees cannot be evaluated empirically, and must be proven --- without…
MAMBA learns policies competitive with multiple conflicting oracles.
problem Learning policies from multiple conflicting oracles in reinforcement learning.
method MAMBA uses a gradient estimator in the style of GAE to optimize policies, leveraging demonstrations from multiple weak oracles.
result MAMBA outperforms the state-of-the-art in learning policies competitive with multiple conflicting oracles.
A major challenge in contextual bandits is to design general-purpose algorithms that are both practically useful and theoretically well-founded. We present a new technique that has the empirical and computational advantages of realizability-based approaches combined with the flexibility of agnostic methods. Our algorit…
Three new oracle-efficient algorithms for private synthetic data release.
problem Constructing private synthetic data that preserves statistical query answers.
method Oracle-efficient algorithms using optimization oracles for differential privacy.
result Better accuracy in large workload and high privacy regime compared to state-of-the-art.
RPI combines imitation and reinforcement learning to improve policies efficiently.
problem High sample complexity in reinforcement learning.
method Active interleaving between imitation and reinforcement learning, using oracle queries for exploration.
result RPI outperforms existing methods across various domains.
Study on tradeoffs between mistakes and ERM oracle calls in online and transductive learning.
problem Analyzing online and transductive learning with limited ERM and weak consistency oracle access.
method Proves lower bounds and upper bounds on mistakes and oracle calls, considering realizable and agnostic cases.
result Achieves optimal mistake bounds with weak consistency queries for certain concept classes.
Extracts fairness truth from classifiers using an oracle.
problem Fairness in classification algorithms without context.
method Uses fairness oracle to learn underlying fairness truth.
result Extracts metric fairness from classifiers.
New algorithm finds k-centers from noisy distance estimates.
problem Finding k-centers in unknown metric spaces with noisy distance queries.
method Active algorithms using UCB, Thompson Sampling, and Track-and-Stop.
result Approximation ratio of two with high probability.
Algorithm identifies nearest mode in noisy data.
problem Identifying the point with the minimum k-th nearest neighbor distance in unknown multivariate probability density.
method Sequential learning algorithm using noisy oracle queries to adaptively decide which points to query.
result Upper bounds on query complexity show significant improvement over baselines.
Study validates metrics for offline MBO using diffusion models.
problem Evaluate metrics for offline MBO without ground truth oracle.
method Propose and quantify validation metrics over datasets.
result Identify most effective validation metrics.
A new hybrid algorithm reduces stochastic gradient evaluations for nonconvex optimization.
problem Solving stochastic composite nonconvex optimization problems efficiently.
method Proposes a new hybrid variance-reduced proximal gradient method with a stochastic gradient estimator.
result Achieves optimal stochastic oracle complexity bound with one less gradient evaluation.
Oracle-efficient algorithm for offline RL with partial data coverage.
problem Offline reinforcement learning with partial data coverage and constraints.
method PDOCRL, a primal-dual algorithm with decomposed linear-programming formulation.
result Near-optimal, near-feasible policy with \(\widetilde{\mathcal O}(ε^{-2})\) sample guarantee.
One of the most effective algorithms for differentially private learning and optimization is objective perturbation. This technique augments a given optimization problem (e.g. deriving from an ERM problem) with a random linear term, and then exactly solves it. However, to date, analyses of this approach crucially rely …
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.
New algorithm reduces super-arm selection complexity exponentially.
problem Combinatorial multi-armed bandits with cardinality constraint.
method Combination of group-testing and quantized Thompson sampling.
result Achieves same regret order as state-of-the-art algorithms with perfect oracle, but with reduced complexity.
High throughput genetic sequencing arrays with thousands of measurements per sample and a great amount of related censored clinical data have increased demanding need for better measurement specific model selection. In this paper we establish strong oracle properties of nonconcave penalized methods for nonpolynomial (N…
New algorithm solves complex optimization problems without needing projections.
problem Optimizing nested functions under convex constraints with noisy evaluations.
method Projection-free conditional gradient-type algorithm for smooth stochastic multi-level composition optimization.
result The algorithm achieves ε ε ε -stationary solutions with complexity bounds independent of ε ε ε and T T T . New algorithm optimizes convex functions with noisy evaluations in one dimension.
problem Optimizing convex functions with noisy zero-order evaluations in one dimension.
method Proposed a computationally efficient algorithm achieving O ( 1 / T ) O(1/\sqrt{T}) O ( 1/ T ) convergence rate. result Achieved the optimal O ( 1 / T ) O(1/\sqrt{T}) O ( 1/ T ) convergence rate, closing the gap in one dimension. Propose an XMSE-aware mixed estimator for EB that interpolates between ML and EB shrinkage.
problem Kernel-based EB estimation may be worse than ML when the kernel is poorly aligned with the true parameter.
method An XMSE-aware mixed estimator that interpolates between ML and EB shrinkage.
result Fixed-weight XMSE is a scalar quadratic, yielding a closed-form oracle mixing weight that is no worse than both ML and the base EB estimator at the XMSE scale.
New algorithm optimally evaluates policies with linear approximations.
problem Policy evaluation with linear function approximation.
method Accelerated, variance-reduced fast temporal difference algorithm (VRFTD).
result VRFTD matches both deterministic and stochastic lower bounds.
Method selects best estimator for off-policy evaluation.
problem Choosing the best estimator for off-policy evaluation.
method Generic data-driven method for estimator selection.
result Method is competitive with oracle estimator, up to a constant factor.
Improved cutting plane method for convex optimization and games.
problem Efficiently finding points in convex sets or proving they do not contain balls.
method Optimal cutting plane algorithm using leverage scores and advanced data structures.
result Significant improvement in time complexity for convex optimization and games.
New algorithm reduces ERM problem size while maintaining accuracy.
problem Empirical risk minimization problem size reduction.
method Adaptive Deterministic Uniform-Weight Trimming (ADUWT) algorithm.
result Uniform ( 1 ± ε ) (1\pm\varepsilon) ( 1 ± ε ) relative-error approximation for ERM objective. New method improves transductive learning predictions with multiplicative oracle inequalities.
problem Improving transductive learning predictions with known covariates.
method Median of Level-Set Aggregation (MLSA) for transductive LOO prediction.
result Proved multiplicative oracle inequality for LOO error.
Paper tackles noisy comparison oracle for robust clustering algorithms.
problem Finding robust clustering algorithms under noisy comparison oracle.
method Develops algorithms for k-center clustering and agglomerative hierarchical clustering using noisy comparison oracle.
result Proves robust algorithms achieve good approximation guarantees with high probability.
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.
Sufficient supervised information is crucial for any machine learning models to boost performance. However, labeling data is expensive and sometimes difficult to obtain. Active learning is an approach to acquire annotations for data from a human oracle by selecting informative samples with a high probability to enhance…
QActor optimizes learning from noisy labeled data streams by querying experts for clean labels.
problem Learning from noisy labeled data in continuous streams with limited oracle queries.
method Combines quality models for filtering and oracle queries for true labels, dynamically adjusting query limits.
result QActor nearly matches optimal accuracy with up to 6% additional ground truth data from experts.
It is not easy to design and run Convolutional Neural Networks (CNNs) due to: 1) finding the optimal number of filters (i.e., the width) at each layer is tricky, given an architecture; and 2) the computational intensity of CNNs impedes the deployment on computationally limited devices. Oracle Pruning is designed to rem…
PCTS optimizes noisy, delayed, multi-fidelity feedbacks in black-box optimization.
problem Optimizing unknown functions with noisy, delayed, and multi-fidelity feedbacks.
method ProCrastinated Tree Search (PCTS) with DUCB1 and DUCBV algorithms.
result PCTS achieves better regret bounds for delayed, noisy, and multi-fidelity feedbacks.
Oracle-efficient algorithms reduce combinatorial semi-bandit regret to logarithmic time.
problem Scalability issue in combinatorial semi-bandit problems due to high combinatorial optimization costs.
method Oracle-efficient frameworks that minimize oracle queries while maintaining tight regret guarantees.
result Achieved i l d e O ( T ) ilde{O}(\sqrt{T}) i l d e O ( T ) regret with O ( log log T ) O(\log\log T) O ( log log T ) oracle queries for worst-case linear rewards. New analysis shows Thompson Sampling can work with greedy approximations in combinatorial bandits.
problem Thompson Sampling's theoretical limits with greedy approximations in combinatorial semi-bandits.
method Study with greedy oracle, providing lower and upper bounds on regret.
result First theoretical results showing TS can work with greedy approximations, breaking misconceptions.
New algorithms sample convex bodies using Markov chains and restricted Gaussian oracles.
problem Sampling uniformly from convex bodies efficiently.
method Markov chain Monte Carlo with proximal sampler and restricted Gaussian oracle.
result Efficient implementation of RGO for uniform sampling on convex bodies.
Minimax PAC bounds for learning in exogenous contextual MDPs
problem PAC learning in tabular discounted Markov decision processes with exogenous i.i.d. contexts
method Variance-reduced algorithm for policy evaluation, best-value estimation, and best-policy extraction
result Minimax optimal sample complexity
Algorithm samples from composite log-concave distributions using gradient evaluations and restricted Gaussian oracles.
problem Sampling from composite log-concave distributions with limited gradient evaluations.
method Proximal gradient algorithm with RGO for g g g and strong/strongly convex conditions for f f f . result Achieves ε ε ε error in total variation distance in O ~ ( κ d log 4 ( 1 / ε ) ) \widetilde{\mathcal O}(κ\sqrt d \log^4(1/ε)) O ( κ d log 4 ( 1/ ε )) iterations. We study the problem of interactively learning a binary classifier using noisy labeling and pairwise comparison oracles, where the comparison oracle answers which one in the given two instances is more likely to be positive. Learning from such oracles has multiple applications where obtaining direct labels is harder bu…
Study examines how uncertainty visualization affects analyst trust in automated classification systems.
problem The impact of uncertainty on analyst trust in automated classification systems.
method Empirical study evaluating different active learning query policies and visualizations.
result Query policy significantly influences analyst trust in automated classification systems.
Paper tackles dynamic pricing in a geometrically decaying environment, achieving better occupancy with lower rates.
problem Minimizing expected loss in a dynamically changing environment with decisions dependent on the data distribution.
method Introduces algorithms for information and loss function settings, using repeated decision deployment to allow mixing of the environment.
result Iteration complexity matches first and zero order stochastic gradient methods up to logarithmic factors.
VR-GHAL method solves stochastic fixed-point equations with high probability.
problem Solving stochastic fixed-point equations in normed spaces with nonexpansive or contractive operators.
method VR-GHAL, a variance-reduced gradual Halpern method for quadratically smoothable Banach spaces, using clipped stochastic differences.
result The method achieves a high-probability residual bound, reducing the residual nearly geometrically across epochs.
New oracle uses uncertainty for active classification with noisy feedback.
problem Improving query complexity in interactive binary classifier learning.
method Proposes a new pairwise comparison oracle that considers uncertainty and an adaptive labeling algorithm.
result Demonstrates improved performance and efficiency compared to existing methods.
Quantum oracles help identify counterfactuals better than classical ones.
problem Identifying unknown causal parameters in causal models.
method Using quantum oracles to query and identify all causal parameters and counterfactuals.
result Quantum oracles enable identification of all two-way joint counterfactuals and tighter bounds on higher-order counterfactuals.
SoQal reduces oracle label requests in active learning by up to 35%.
problem Exploiting unlabelled data in healthcare requires costly oracle labeling.
method Dynamic questioning strategy to minimize oracle label requests.
result SoQal reduces oracle label requests by up to 35%.
Paper addresses online alignment of large language models under uncertain preference feedback.
problem Online alignment of large language models with misspecified preference feedback.
method Formulates an oracle-robust objective as a worst-case optimization problem for log-linear policies, and develops projected stochastic composite updates.
result Shows that the robust objective admits an exact closed-form decomposition and achieves O ~ ( ε − 2 ) \widetilde{O}(\varepsilon^{-2}) O ( ε − 2 ) oracle complexity. We consider the problem of minimizing the sum of submodular set functions assuming minimization oracles of each summand function. Most existing approaches reformulate the problem as the convex minimization of the sum of the corresponding Lovász extensions and the squared Euclidean norm, leading to algorithms requiring …