ReLax: An Efficient and Scalable Recourse Explanation Benchmarking Library using JAX
TL;DR: We propose ReLax, a JAX-based benchmarking library which is designed for efficient and scalable recourse explanations.
Abstract: Despite the progress made in the field of algorithmic recourse, current research practices remain constrained, largely restricting benchmarking and evaluation of recourse methods to medium-sized datasets (approximately 50k data points) due to the severe runtime overhead of recourse generation. This constraint impedes the pace of research development in algorithmic recourse and raises concerns about the scalability of existing methods. To mitigate this problem, we propose ReLax, a JAX-based benchmarking library, designed for efficient and scalable recourse explanations. ReLax supports a wide range of recourse methods and datasets and offers performance improvements of at least two orders of magnitude over existing libraries. Notably, we demonstrate that ReLax is capable of benchmarking real-world datasets of up to 10M data points, roughly 200 times the scale of current practices, without imposing prohibitive computational costs. ReLax is fully open-sourced and can be accessed at https://github.com/BirkhoffG/jax-relax.
Submission Track: Full Paper Track
Application Domain: None of the above / Not applicable
Clarify Domain: We propose a benchmarking library for recourse and counterfactual explanations.
Survey Question 1: we propose ReLax, a JAX-based benchmarking library, designed for efficient and scalable recourse explanations. Notably, we demonstrate that ReLax is capable of benchmarking real-world datasets of up to 10M data points, roughly 200 times the scale of current research practices, without imposing prohibitive computational costs.
Survey Question 2: Not applicable as my work focuses on developing explainable AI methods.
Survey Question 3: Not applicable as my work focuses on developing explainable AI methods.
Submission Number: 45
Loading