Problem
Research questions and friction points this paper is trying to address.
many-objective optimization
preference learning
heterogeneous preferences
trade-off space
human value structures
Mixture-Model Preference Learning for Many-Objective Bayesian Optimization
🤖 AI Summary
This work addresses the challenges of trade-off space explosion and the heterogeneity and context-dependence of human preferences in multi-objective preference optimization by proposing a Bayesian framework that employs a Dirichlet process mixture model to automatically infer a small set of latent preference prototypes. The approach jointly models uncertainty in both the prototypes and their associated weights, innovatively replacing the conventional assumption of a single utility function with a mixture of preference prototypes. Integrated with a hybrid active querying strategy and theoretical regret bounds, the method efficiently acquires informative feedback while uncovering complex preference structures often missed by standard metrics. Experimental results demonstrate that the proposed method significantly outperforms existing baselines on both synthetic and real-world multi-objective benchmarks, with diagnostic analyses successfully recovering hidden preference patterns.
📝 Abstract
Preference-based many-objective optimization faces two obstacles: an expanding space of trade-offs and heterogeneous, context-dependent human value structures. Towards this, we propose a Bayesian framework that learns a small set of latent preference archetypes rather than assuming a single fixed utility function, modelling them as components of a Dirichlet-process mixture with uncertainty over both archetypes and their weights. To query efficiently, we designing hybrid queries that target information about (i) mode identity and (ii) within-mode trade-offs. Under mild assumptions, we provide a simple regret guarantee for the resulting mixture-aware Bayesian optimization procedure. Empirically, our method outperforms standard baselines on synthetic and real-world many-objective benchmarks, and mixture-aware diagnostics reveal structure that regret alone fails to capture.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Mixture-Model Preference Learning
Many-Objective Bayesian Optimization
Dirichlet-Process Mixture
Latent Preference Archetypes
Hybrid Preference Queries
🔎 Similar Papers
No similar papers found.
