Scalable and Stable Surrogates for Flexible Classifiers with Fairness Constraints
Authors: Henry C Bendekgey, Erik Sudderth
NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our surrogates perform comparably to the state-of-the-art on low-dimensional fairness benchmarks, while achieving superior accuracy and stability for more complex computer vision and natural language processing tasks.We compare relaxations by training fairness-regularized logistic regression on tabular data, and fairness-regularized deep neural networks on image and text data. |
| Researcher Affiliation | Academia | Harry Bendekgey hbendekg@uci.edu Erik B. Sudderth sudderth@uci.edu Department of Computer Science, University of California Irvine School of Information and Computer Science, Irvine, CA, USA |
| Pseudocode | No | The paper describes mathematical formulations and algorithmic steps in prose and equations but does not contain a formally labeled "Pseudocode" or "Algorithm" block. |
| Open Source Code | No | The paper does not provide any explicit statement about releasing source code or a link to a code repository for the methodology described. |
| Open Datasets | Yes | Adult. The Adult data set [36] is one of the most popular in the fair classification literature...,COMPAS. The COMPAS data set was compiled by Angwin et al. [2]...,Celeb A. The Celeb Faces Attributes data set [39]...,Faces of the World (FOTW) data set [20]...,Yelp Text Data. We use a subset of the Yelp review data set [13]... |
| Dataset Splits | No | The paper mentions "training data" and "test data" in figures and text but does not explicitly provide specific train/validation/test split percentages, sample counts, or detailed methodology for creating these splits in the main body. It states "Methods for hyperparameter selection, data pre-processing, and optimization are detailed in the supplement", which might contain this information, but it's not present in the main text. |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., CPU, GPU models, memory, or cloud instances) used for running its experiments. |
| Software Dependencies | No | The paper mentions software components like 'WRN-50-2' and 'Google’s transformer-based BERT' but does not specify their version numbers or the versions of other ancillary software (e.g., deep learning frameworks, programming languages) used in the experiments. |
| Experiment Setup | No | The paper explicitly states: 'Methods for hyperparameter selection, data pre-processing, and optimization are detailed in the supplement.' This indicates that specific experimental setup details are not provided in the main text. |