Disentangling Revealed Preference From Rationalization by a Preference
The weak axiom of revealed preference (WARP) ensures that the revealed preference (i) is a preference relation (i.e., it is complete and transitive) and (ii) rationalizes the choices. However, when WARP fails, either one of these two properties is violated, but it is unclear which one it is. We provide an alternative characterization of WARP by showing that WARP is equivalent to the conjunction of two axioms each of which separately guarantees (i) and (ii).
Code (0)
등록된 구현이 없습니다.
Tasks
RelationSimilar Papers 제목 키워드 기반
Mind the Gap: How Elicitation Protocols Shape the Stated-Revealed Preference Gap in Language Models
Recent work identifies a stated-revealed (SvR) preference gap in language models (LMs): a mismatch between the values models endorse and the choices they make in context. Existing evaluations rely heavily on binary force…
An algebraic approach to revealed preferences
We propose and develop an algebraic approach to revealed preference. Our approach dispenses with non algebraic structure, such as topological assumptions. We provide algebraic axioms of revealed preference that subsume p…
A General Revealed Preference Test for Quasilinear Preferences: Theory and Experiments
We provide a generalized revealed preference test for quasilinear preferences. The test applies to nonlinear budget sets and non-convex preferences as those found in taxation and nonlinear pricing contexts. We study the …
LiteraryTaste: A Preference Dataset for Creative Writing Personalization
People have different creative writing preferences, and large language models (LLMs) for these tasks can benefit from adapting to each user's preferences. However, these models are often trained over a dataset that consi…
Disentangling Preference Representation and Text Generation for Efficient Individual Preference Alignment
Aligning Large Language Models (LLMs) with general human preferences has been proved crucial in improving the interaction quality between LLMs and human. However, human values are inherently diverse among different indiv…
Text Generation