Meta Optimality for Demographic Parity Constrained Regression via Post-Processing
Authors: Kazuto Fukuchi
ICML 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | In this paper, we provide meta-theorems that can be applied to various situations to validate the fair minimax optimality of the corresponding regression algorithms. Furthermore, we demonstrate that fair minimax optimal regression can be achieved through post-processing methods... Our main result is a meta-theorem that characterizes the fair minimax optimal error (see Equation (1)), showing how its convergence rate depends on P. We begin by introducing several technical assumptions on P before presenting our main meta-theorem. |
| Researcher Affiliation | Academia | 1Institute of Systems and Information Engineering, University of Tsukuba, Ibaraki, Japan 2RIKEN AIP, Tokyo, Japan. Correspondence to: Kazuto Fukuchi <EMAIL>. |
| Pseudocode | Yes | Algorithm 1 Optimal fair regression input Samples (Y (s) 1 , X(s) 1 ), ..., (Y (s) 2n , X(s) 2ns) output fn,: 1: Construct fn,: using a minimax optimal conventional regression algorithm with halves of samples (Y (s) 1 , X(s) 1 ), ..., (Y (s) ns , X(s) ns ) 2: Construct ϑn,: by Equation (5) with the remaining samples fn,s(X(s) n+1), ..., fn,s(X(s) 2ns)) 3: fn,s(x) = (ϑn,s fn,s)(x) |
| Open Source Code | No | The paper does not provide any statement or link regarding the release of open-source code for the described methodology. |
| Open Datasets | No | The paper is theoretical and focuses on algorithm design and convergence rate analysis for fair regression. It does not conduct experiments on specific datasets or provide access information for any open datasets. |
| Dataset Splits | No | The paper is theoretical and does not describe experiments using specific datasets with training/test/validation splits. It discusses 'samples' in a general, theoretical context (e.g., 'Given samples consisting of ns i.i.d. copies of (X(s), Y (s))'). |
| Hardware Specification | No | The paper is theoretical and focuses on mathematical proofs and algorithm design. It does not include an experimental section or specify any hardware used for running experiments. |
| Software Dependencies | No | The paper is theoretical, presenting meta-theorems and algorithms without detailing an implementation. Therefore, no specific software dependencies with version numbers are mentioned. |
| Experiment Setup | No | The paper is theoretical, focusing on meta-theorems, algorithms, and convergence rates. It does not include an experimental section with specific setup details such as hyperparameters or training configurations. |