Bias Detection via Signaling

Authors: Yiling Chen, Tao Lin, Ariel D. Procaccia, Aaditya Ramdas, Itai Shapira

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Our goal in this paper is to develop algorithms that are able to detect bias in the form of non-Bayesian updating of beliefs. To our knowledge, we are the first to formalize and analytically address this problem, and we aim to build an initial framework that future work would build on. and Our main results include a characterization of scenarios where a single signal suffices and a computationally efficient algorithm to compute optimal signaling schemes.
Researcher Affiliation Academia Yiling Chen Harvard University yiling@seas.harvard.edu, Tao Lin Harvard University tlin@g.harvard.edu, Ariel D. Procaccia Harvard University arielpro@g.harvard.edu, Aaditya Ramdas Carnegie Mellon University aramdas@cmu.edu, Itai Shapira Harvard University itaishapira@g.harvard.edu
Pseudocode Yes Algorithm 1: Linear program to compute the optimal signaling scheme
Open Source Code No The NeurIPS Paper Checklist indicates 'NA' for 'Does the paper provide open access to the data and code', meaning no code is provided.
Open Datasets No The information is insufficient. The paper is theoretical and does not conduct empirical studies with datasets. The NeurIPS Paper Checklist explicitly states 'NA' for 'Does the paper fully disclose all the information needed to reproduce the main experimental results of the paper'.
Dataset Splits No The information is insufficient. The paper is theoretical and does not involve empirical studies with dataset splits. The NeurIPS Paper Checklist explicitly states 'NA' for questions related to experimental reproducibility.
Hardware Specification No The information is insufficient. The paper is theoretical and does not describe any specific hardware used for experiments.
Software Dependencies No The information is insufficient. The paper is theoretical and does not specify any software dependencies with version numbers for experimental reproducibility.
Experiment Setup No The information is insufficient. The paper is theoretical and does not describe an experimental setup with specific hyperparameters or training configurations.