Optimal Design for Human Preference Elicitation

Abstract

Learning of preference models from human feedback has been central to recent advances in artificial intelligence. Motivated by the cost of obtaining high-quality human annotations, we study efficient human preference elicitation for learning preference models. The key idea in our work is to generalize optimal designs, an approach to computing optimal information-gathering policies, to lists of items that represent potential questions with answers. The policy is a distribution over the lists and we elicit preferences from them proportionally to their probabilities. To show the generality of our ideas, we study both absolute and ranking feedback models on items in the list. We design efficient algorithms for both and analyze them. Finally, we demonstrate that our algorithms are practical by evaluating them on existing question-answering problems.

Cite

Text

Mukherjee et al. "Optimal Design for Human Preference Elicitation." Neural Information Processing Systems, 2024. doi:10.52202/079017-2861

Markdown

[Mukherjee et al. "Optimal Design for Human Preference Elicitation." Neural Information Processing Systems, 2024.](https://mlanthology.org/neurips/2024/mukherjee2024neurips-optimal/) doi:10.52202/079017-2861

BibTeX

@inproceedings{mukherjee2024neurips-optimal,
  title     = {{Optimal Design for Human Preference Elicitation}},
  author    = {Mukherjee, Subhojyoti and Lalitha, Anusha and Kalantari, Kousha and Deshmukh, Aniket and Liu, Ge and Ma, Yifei and Kveton, Branislav},
  booktitle = {Neural Information Processing Systems},
  year      = {2024},
  doi       = {10.52202/079017-2861},
  url       = {https://mlanthology.org/neurips/2024/mukherjee2024neurips-optimal/}
}