arXiv research

New methods estimate probabilities from pairwise comparisons, adapting to difficulty.

New algorithm learns HMM from pairwise co-occurrences, improving topic modeling.

Active seriation recovers item order from noisy pairwise similarity measurements.

The paper proposes a method to calibrate evidential clustering using bootstrapped finite mixture models.

Efficiently approximates softmax probabilities for large-scale inference.

Improved model capacity for graph cut algorithms by relaxing submodularity constraints.

This work proposes a new method to estimate joint probability from pairwise marginals, reducing sample complexity.

An algorithm ranks items based on noisy comparisons, proving optimality without parametric assumptions.

New model for pairwise comparisons without stochastic transitivity.

This paper examines the problem of ranking a collection of objects using pairwise comparisons (rankings of two objects). In general, the ranking of $n$ objects can be identified by standard sorting methods using $n log_2 n$ pairwise comparisons. We are interested in natural situations in which relationships among the o…

The aim of this short note is to draw attention to a method by which the partition function and marginal probabilities for a certain class of random fields on complete graphs can be computed in polynomial time. This class includes Ising models with homogeneous pairwise potentials but arbitrary (inhomogeneous) unary pot…

Sampling a fraction of pairs can match full evaluation in machine learning losses.

Study on deep neural networks for reward modeling with pairwise comparison data.

Develops a statistical framework to measure uncertainty in model rankings based on human preferences.

New model accounts for scale variation and noise in pairwise comparisons.

LOT Wassmap speeds up Wasserstein space manifold learning.

Two-hop walks reveal PageRank order in networks.

A robust ranking algorithm outperforms previous methods in efficiency and guarantees.

There are various parametric models for analyzing pairwise comparison data, including the Bradley-Terry-Luce (BTL) and Thurstone models, but their reliance on strong parametric assumptions is limiting. In this work, we study a flexible model for pairwise comparisons, under which the probabilities of outcomes are requir…

An algorithm finds approximate rankings from pairwise comparisons with near-optimal comparisons.

Pairwise quantile regression tackles similarity scoring in biometric systems.

Pairwise learning usually refers to a learning task which involves a loss function depending on pairs of examples, among which most notable ones include ranking, metric learning and AUC maximization. In this paper, we study an online algorithm for pairwise learning with a least-square loss function in an unconstrained …

New framework estimates treatment effects based on preferences.

Bayesian model infers strengths from noisy tennis match outcomes.

Surv-IPTB estimates patient-specific treatment benefits using survival data.

Exact simulation of correlated binary outcomes using PMF constraints and linear programming.

Paper introduces a new method for learning with distributions using dissimilarity measures.

A Markov Chain approach for aligning generative models from pairwise human preferences.

New models for analyzing microbiome data with interactions.

Estimates latent positions in 1D torus from noisy pairwise affinities.

The Viterbi process can be extended indefinitely in a pairwise Markov model.

New model captures intransitive preferences without concave likelihood.

EBBS integrates expert assessments into MIO best-subsets problem.

The paper develops a method to estimate consumer preferences from observed rankings.

Paper tackles learning mixture of RUMs from partial data.

We present a global optimization algorithm for clustering data given the ratio of likelihoods that each pair of data points is in the same cluster or in different clusters. To define a clustering solution in terms of pairwise relationships, a necessary and sufficient condition is that belonging to the same cluster sati…

SyncRank recovers global ranking from noisy comparisons with theoretical guarantees.

This paper optimizes the number of comparisons needed to find the best k items from pairwise comparisons.

Paper studies how to cluster data with a weak oracle, reducing the number of queries needed.

If L_1 and L_2 are two Brunnian links with all pairwise linking numbers 0, then we show that L_1 and L_2 are equivalent if and only if they have homeomorphic complements. In particular, this holds for all Brunnian links with at least three components. If L_1 is a Brunnian link with all pairwise linking numbers 0, and t…

The paper studies ranking algorithms from pairwise and listwise comparisons, deriving lower bounds and optimal algorithms.

Optimizes identifying top-k items from comparisons with minimal comparisons.

A new model for supervised learning to rank using gradient estimation.

New estimators show consistent estimation is possible with randomized item assignment.

The problem of bipartite ranking, where instances are labeled positive or negative and the goal is to learn a scoring function that minimizes the probability of mis-ranking a pair of positive and negative instances (or equivalently, that maximizes the area under the ROC curve), has been widely studied in recent years. …

A new algorithm optimizes multi-way matching problems.

Distance-based hierarchical clustering (HC) methods are widely used in unsupervised data analysis but few authors take account of uncertainty in the distance data. We incorporate a statistical model of the uncertainty through corruption or noise in the pairwise distances and investigate the problem of estimating the HC…

Proposes a new method to improve multiclass probability calibration.