Optimizes recommender selection online with D-optimal design.
problem Finding the optimal recommender in online exploration-exploitation.
method Leverages D-optimal design from statistics to maximize information gain.
result Achieves maximum information gain during online exploration.
Experimental design is a classical statistics problem and its aim is to estimate an unknown m-dimensional vector β from linear measurements where a Gaussian noise is introduced in each measurement. For the combinatorial experimental design problem, the goal is to pick k out of the given n experiments so as to m…
A new algorithm selects data subsets avoiding outliers and high leverage points.
problem Outliers and high leverage points skew model estimates in subsamples.
method Unsupervised and supervised exchange procedures to select nearly D-optimal subsets.
result The new methods improve model accuracy by avoiding influential points.
A2D2E designs optimal evaluations for estimating main effects in black-box models.
problem Sensitivity to OOD evaluations and instability under feature correlation in estimating black-box model effects.
method A2D2E uses D-optimal hypercube designs to minimize main effect estimation variance.
result A2D2E outperforms existing methods, especially under high feature correlation.
New algorithm learns human preferences from few comparisons efficiently.
problem Learning human preferences from limited comparison feedback.
method Formulated as D-optimal design for Plackett-Luce model, solved using randomized Frank-Wolfe algorithm.
result Proposed algorithm efficiently solves D-optimal design problem for Plackett-Luce objective.
Enhanced PC2 improves surrogate modeling for high-dimensional problems.
problem Degrading performance and efficiency of PC2 in high-dimensional parameter spaces. method Integrates SULM solver and D-optimal sampling strategy into PC2 framework. result Enhanced PC2 demonstrates better comprehensive capability and efficiency. Active learning framework for optimizing human preferences in reinforcement learning.
problem Selecting most informative feedback for training models of human preferences.
method Proposes an active learning framework to collect preferential feedback online or offline.
result Errors in DPO logit estimates diminish with more feedback.
Gonogo offers tools for sensitivity experiments in R.
problem Conducting, analyzing, and simulating sensitivity experiments.
method Suite of R functions for various adaptive procedures.
result Achieving overlapping data and refining testing in distribution tails.
Algorithm reduces online regret by leveraging offline data in linear bandits.
problem Online regret minimization in linear bandits with offline data.
method OOPE algorithm using extended D-optimal design.
result Substantial reduction in online regret compared to prior work.
Optimal sampling reduces power grid data analysis costs.
problem Efficient online analysis of high-speed, correlated IoT data.
method D-optimality criterion-based sampling methods combining Bernoulli and leverage score sampling.
result Leverage score sampling improves computational efficiency and outperforms benchmarks.
Online active learning improves model performance in high-volume production processes.
problem Outliers degrade the predictive performance of online active learning models.
method Proposes a robust estimator and bounds the search area of a conditional D-optimal algorithm.
result The proposed method improves the performance of online active learning in the presence of outliers.
We demonstrate how to scalably solve a class of constrained self-concordant minimization problems using linear minimization oracles (LMO) over the constraint set. We prove that the number of LMO calls of our method is nearly the same as that of the Frank-Wolfe method in the L-smooth case. Specifically, our Newton Frank…
In this paper, we consider an online optimization process, where the objective functions are not convex (nor concave) but instead belong to a broad class of continuous submodular functions. We first propose a variant of the Frank-Wolfe algorithm that has access to the full gradient of the objective functions. We show t…
New active learning methods for Gaussian process improve predictive modeling of composite fuselage.
problem Improving predictive modeling of composite fuselage with limited training samples and uncertainties.
method Proposed two new active learning algorithms for Gaussian process considering uncertainties.
result The proposed approach realizes better prediction performance for automatic shape control of composite fuselage.
In experimental design, we are given n vectors in d dimensions, and our goal is to select k≪n of them to perform expensive measurements, e.g., to obtain labels/responses, for a linear regression task. Many statistical criteria have been proposed for choosing the optimal design, with popular choices including A…
Dual active learning improves RLHF by selecting optimal conversations and teachers.
problem Efficiently aligning LLMs with human preferences using RLHF from feedback.
method Offline RL for conversation and teacher selection, dual active reward learning, pessimistic RL.
result The proposed algorithm achieves minimal generalized variance and outperforms state-of-the-arts.
We study the optimal design problems where the goal is to choose a set of linear measurements to obtain the most accurate estimate of an unknown vector in d dimensions. We study the A-optimal design variant where the objective is to minimize the average variance of the error in the maximum likelihood estimate of th…
Study optimizes sensor placement for accurate parameter estimation in complex systems.
problem Challenges in parameter estimation with limited or noisy data.
method Physics-Informed Neural Networks (PINNs) for optimal sensor placement and parameter estimation.
result PINNs-based framework achieves higher accuracy in parameter estimation compared to random sensor placements.
The paper introduces a framework to select efficient datasets for preserving model rankings.
problem Efficient evaluation of machine learning models on small, representative datasets.
method Bootstrap aggregation, clustering, design criteria, random baselines, and greedy farthest-first (FAFI).
result Several selection strategies improve rank preservation compared to random subsets, especially in time series classification.
ALMAB-DC optimizes expensive black-box experiments using active learning and distributed computing.
problem Efficiently optimizing expensive, gradient-free objectives in computational statistics and machine learning.
method Combines active learning, multi-armed bandits, and distributed asynchronous computing.
result Achieves lower simple regret and superior performance in various tasks compared to non-ALMAB baselines.