In the context of reinforcement learning from human feedback (RLHF), the reward function is generally derived from maximum likelihood estimation of a random utility model based on pairwise comparisons made by humans. The problem of learning a reward function is one of preference aggregation that, we argue, largely falls within the scope of social choice theory. From this perspective, we can evaluate different aggregation methods via established axioms, examining whether these methods meet or fail well-known standards. We demonstrate that both the Bradley-Terry-Luce Model and its broad generalizations fail to meet basic axioms. In response, we develop novel rules for learning reward functions with strong axiomatic guarantees. A key innovation from the standpoint of social choice is that our problem has a linear structure, which greatly restricts the space of feasible rules and leads to a new paradigm that we call linear social choice.

翻译：在基于人类反馈的强化学习（RLHF）中，奖励函数通常源自对基于人类成对比较的随机效用模型进行最大似然估计。学习奖励函数的问题属于偏好聚合问题，我们认为该问题在很大程度上属于社会选择理论的范畴。从这个视角出发，我们可以通过既定的公理来评估不同的聚合方法，检验这些方法是否符合或违背公认的标准。我们证明，Bradley-Terry-Luce模型及其广泛推广均未能满足基本公理。为此，我们提出了具有强公理保证的学习奖励函数的新规则。从社会选择的角度看，一个关键的创新在于我们的问题具有线性结构，这极大地限制了可行规则的空间，并催生了一种我们称之为线性社会选择的新范式。