Develops a statistical framework to measure uncertainty in model rankings based on human preferences.
problem Uncertainty in model rankings based on human preferences due to mismatch between human and model preferences.
method Statistical framework using pairwise comparisons by humans and models to provide rank-sets for each model.
result Rank-sets constructed using only pairwise comparisons by strong models often do not cover the true ranking of human preferences.
A Markov Chain approach for aligning generative models from pairwise human preferences.
problem Aligning generative models from pairwise human preferences.
method Markov Chain from Human Feedback (MCHF)
result MCHF converges geometrically fast to the stationary distribution.
Given a set of pairwise comparisons, the classical ranking problem computes a single ranking that best represents the preferences of all users. In this paper, we study the problem of inferring individual preferences, arising in the context of making personalized recommendations. In particular, we assume that there are …
Bayesian optimization learns DM preferences for multi-outcome experiments.
problem Optimizing expensive experiments with unknown utility functions and multiple outcomes.
method Alternates preference learning and Bayesian optimization, using pairwise comparisons.
result Preference exploration strategies improve Bayesian optimization performance.
We study the problem of ranking a set of items from nonactively chosen pairwise preferences where each item has feature information with it. We propose and characterize a very broad class of preference matrices giving rise to the Feature Low Rank (FLR) model, which subsumes several models ranging from the classic Bradl…
Pref-SHAP explains preferences using Shapley values.
problem Challenging problem of preference explanation in machine learning.
method Shapley value-based model explanation framework for pairwise comparison data.
result Richer and more insightful explanations obtained over baseline.
Best-of-N sampling reveals reward targets from preference data, influencing N and base distribution choices.
problem Understanding reward extraction from Best-of-N preference data and optimal N and base distribution choices.
method Specialized analysis of preference data via induced conditional distribution, deriving reward targets and design principles.
result Reward targets are explicit functions of N and base distribution, and bounded-class minimizers approach these targets as N grows.
Derives a new objective to learn from human preferences without approximations.
problem Learning from human preferences through RLHF relies on approximations that can lead to pitfalls.
method Derives a new general objective ΨPO that bypasses both approximations. result Demonstrates the superiority of the new objective to Direct Preference Optimisation (DPO) empirically.
In this paper we consider the collaborative ranking setting: a pool of users each provides a small number of pairwise preferences between d possible items; from these we need to predict preferences of the users for items they have not yet seen. We do so by fitting a rank r score matrix to the pairwise data, and pro…
The paper designs tests for comparing ranked preference data and finds significant differences.
problem Comparing pairwise comparison and ranking data in various applications.
method Developed two-sample tests for pairwise comparison and ranking data, proving upper and lower bounds.
result Upper and lower bounds show tightness of the proposed tests, and significant differences in preferences were found.
In this paper we present a hybrid active sampling strategy for pairwise preference aggregation, which aims at recovering the underlying rating of the test candidates from sparse and noisy pairwise labelling. Our method employs Bayesian optimization framework and Bradley-Terry model to construct the utility function, th…
Bayesian optimization agent learns user preferences from pairwise comparisons.
problem Learning user preferences from unknown and infinite choices.
method Sequential Bayesian optimization with pairwise comparisons.
result Optimal agent strategy minimizes remaining system uncertainty.
New framework estimates treatment effects based on preferences.
problem Estimating treatment effects with flexible outcomes.
method Preference-based Conditional Treatment Effect (CPTE) framework.
result CPTE provides interpretable targets and new identifiability conditions.
RLHF performs well despite violating social choice theory axioms.
problem RLHF's empirical success contradicts social choice theory axioms.
method Showed RLHF satisfies pairwise majority and Condorcet consistency under mild assumptions, and introduced new alignment criteria.
result RLHF satisfies pairwise majority and Condorcet consistency under mild assumptions, explaining its practical success.
In this paper, we propose an active learning algorithm and models which can gradually learn individual's preference through pairwise comparisons. The active learning scheme aims at finding individual's most preferred choice with minimized number of pairwise comparisons. The pairwise comparisons are encoded into probabi…
The paper explores game-theoretic alignment of LLMs with human preferences, finding limitations and conditions.
problem Aligning LLMs with human preferences using game theory.
method Systematic study of payoff choices in a two-player zero-sum game for desirable alignment properties.
result Impossibility of preference matching in game-theoretic LLM alignment under standard assumptions.
Bayesian optimization with preference learning using monotonic neural networks.
problem Optimizing complex systems with multiple conflicting objectives.
method Proposes a neural network ensemble for utility surrogate modeling, leveraging monotonicity.
result Demonstrates superior performance compared to existing methods.
RCPO uses ranked choice modeling for better LLM alignment.
problem Pairwise preference optimization limits LLM alignment.
method Unified framework combining preference optimization and ranked choice modeling.
result RCPO outperforms competitive baselines in LLM alignment.
A game-theoretic approach to multi-criteria ranking from ordinal data.
problem Ranking objects from ordinal data with multiple criteria.
method Generalizing von Neumann winner to multi-criteria setting using Blackwell's approachability.
result The Blackwell winner can be computed as a convex optimization problem and achieves near-optimal sample complexity.
Study metric learning from limited preference comparisons, showing how low-dimensional structure can still reveal metric information.
problem Learning metric from limited pairwise preference comparisons.
method Ideal point model, divide-and-conquer approach for low-dimensional structure.
result Metric can be jointly identified even with limited comparisons when items exhibit low-dimensional structure.
CnGAN generates synthetic user preferences for non-overlapped users in cross-network recommender systems.
problem Cross-network recommender solutions ignore non-overlapped users, limiting their applicability.
method Multi-task learning, encoder-GAN architecture, user-based pairwise loss function.
result Generated user preferences improve recommendations for non-overlapped users, achieving superior performance.
Modeling preference rankings with salient features to explain irrational choices.
problem Estimating rankings from noisy pairwise comparisons with irrational choices.
method Salient feature preference model with maximum likelihood estimation.
result Strong performance of maximum likelihood estimation on synthetic and real data.
In this paper we investigate two variants of association rules for preference data, Label Ranking Association Rules and Pairwise Association Rules. Label Ranking Association Rules (LRAR) are the equivalent of Class Association Rules (CAR) for the Label Ranking task. In CAR, the consequent is a single class, to which th…
Duel-Evolve uses LLM self-preferences for test-time optimization of discrete outputs.
problem Optimizing LLM outputs at test time with limited or unreliable scalar rewards.
method Duel-Evolve uses pairwise comparisons from the LLM to guide optimization, aggregating them via a Bayesian Bradley-Terry model.
result Achieves significant improvement over existing methods in accuracy.
In multi-objective decision planning and learning, much attention is paid to producing optimal solution sets that contain an optimal policy for every possible user preference profile. We argue that the step that follows, i.e, determining which policy to execute by maximising the user's intrinsic utility function over t…
A new method reduces preference distortion in LLM alignment.
problem Vulnerability of traditional LLM alignment methods to human preference heterogeneity.
method Sign Estimator: A simple, provably consistent, and efficient estimator using binary classification loss.
result Substantially reduces preference distortion over a panel of simulated personas.
Optimizes explanations for better listener understanding.
problem Insufficient consideration of listener preferences in concept-based explanations.
method Iterative training procedure based on direct preference optimization.
result Pragmatic explanations improve both model accuracy and user understanding.
Rank aggregation systems collect ordinal preferences from individuals to produce a global ranking that represents the social preference. Rank-breaking is a common practice to reduce the computational complexity of learning the global ranking. The individual preferences are broken into pairwise comparisons and applied t…
New method detects inconsistencies in AHP matrices using triadic preference reversals.
problem Challenges in assessing consistency in AHP pairwise comparison matrices.
method Triadic preference reversals to detect inconsistencies between pairs of elements.
result 97% accuracy in detecting inconsistencies, significantly surpassing traditional methods.
A fast algorithm speeds up training of pairwise kernels.
problem Training pairwise kernels efficiently for large datasets.
method Generalized vec trick for Kronecker product kernels.
result Pairwise kernels can be expressed as sums of Kronecker products.
We propose a scalable Bayesian preference learning method for jointly predicting the preferences of individuals as well as the consensus of a crowd from pairwise labels. Peoples' opinions often differ greatly, making it difficult to predict their preferences from small amounts of personal data. Individual biases also m…
Robot motions in the presence of humans should not only be feasible and safe, but also conform to human preferences. This, however, requires user feedback on the robot's behavior. In this work, we propose a novel approach to leverage the user's brain signals as a feedback modality in order to decode the judgment of rob…
The paper addresses privacy in rank aggregation using randomized responses.
problem Preserving privacy while aggregating pairwise rankings.
method Adaptive debiasing method for randomized response rankings.
result Established minimax rates for estimation errors and optimal privacy guarantees.
There is a growing need for discrete choice models that account for the complex nature of human choices, escaping traditional behavioral assumptions such as the transitivity of pairwise preferences. Recently, several parametric models of intransitive comparisons have been proposed, but in all cases the maximum likeliho…
The paper develops a method to estimate consumer preferences from observed rankings.
problem Estimating consumer preferences from partial ranking information.
method Interpreting observed rankings as pairwise comparisons, modeling latent utility, and correcting for selection bias.
result The method improves recommendation performance, especially for previously unconsumed products.
Unified framework for aligning LLMs from human feedback.
problem Lack of strong theoretical justification for RLHF and inability to compare methods.
method Reframed alignment as distribution learning from pairwise preferences, proposing three principled objectives.
result Proposed objectives achieve strong non-asymptotic convergence to target LM.
Paper investigates monotonicity issues in AI preference learning.
problem AI models may violate monotonicity when learning preferences.
method Investigates root causes of non-monotonicity in comparison-based preference learning.
result Proves local pairwise monotonicity under mild assumptions.
Automates debiasing for large language model evaluations through Fisher random walk.
problem Rigorous and scalable evaluation of large language models.
method Semiparametric efficient estimator using Fisher random walk for weighted residual balancing.
result Efficient estimation of contextual preference scores for large language models.
We propose a novel parameterized family of Mixed Membership Mallows Models (M4) to account for variability in pairwise comparisons generated by a heterogeneous population of noisy and inconsistent users. M4 models individual preferences as a user-specific probabilistic mixture of shared latent Mallows components. Our k…
Paper improves PBO using Skew Gaussian Processes for better optimization.
problem Optimizing with preference judgments, especially in A/B tests and recommender systems.
method Uses Skew Gaussian Processes to model preference function and exact posterior inference.
result Exact SkewGP posterior leads to better optimization results than Laplace approximation.
We propose a topic modeling approach to the prediction of preferences in pairwise comparisons. We develop a new generative model for pairwise comparisons that accounts for multiple shared latent rankings that are prevalent in a population of users. This new model also captures inconsistent user behavior in a natural wa…
SetRank tackles collaborative ranking from implicit feedback using setwise Bayesian approach.
problem Challenges in pairwise and listwise approaches for implicit feedback.
method SetRank is a novel setwise Bayesian approach that accommodates implicit feedback characteristics.
result SetRank outperforms state-of-the-art baselines on real-world datasets.
Bayesian method helps decision-makers find preferred solutions in multi-objective optimization.
problem Identifying preferred solutions from the Pareto set in multi-objective optimization problems.
method Bayesian model to estimate decision-maker's utility function based on pairwise comparisons, guided by a principled elicitation strategy.
result Superior performance in finding high-utility solutions with a small number of queries.
We consider the problem of probably approximately correct (PAC) ranking n items by adaptively eliciting subset-wise preference feedback. At each round, the learner chooses a subset of k items and observes stochastic feedback indicating preference information of the winner (most preferred) item of the chosen subset …
We consider the predictive problem of supervised ranking, where the task is to rank sets of candidate items returned in response to queries. Although there exist statistical procedures that come with guarantees of consistency in this setting, these procedures require that individuals provide a complete ranking of all i…
Active learning improves ordering of items with contextual attributes.
problem Learning accurate item orderings from pairwise comparisons, especially when exhaustive comparisons are impractical.
method Proposes an active learning strategy that samples items to minimize expected ordering error, accounting for uncertainty in comparisons.
result Superior sample efficiency and generalization compared to non-contextual ranking approaches and active preference learning baselines.
In this paper, we propose a novel ranking framework for collaborative filtering with the overall aim of learning user preferences over items by minimizing a pairwise ranking loss. We show the minimization problem involves dependent random variables and provide a theoretical analysis by proving the consistency of the em…
Bayesian optimization with preference learning identifies preferred solutions in multi-objective problems.
problem Optimizing multiple criteria with decision maker preferences in expensive functions.
method Bayesian optimization with interactive preference learning and active acquisition function.
result Identifies the most preferred solution with reduced interaction cost.