# AP-Perf: Incorporating Generic Performance Metrics in Differentiable Learning

arXiv: https://arxiv.org/abs/1912.00965 · HTML: https://mlsanity.com/paper/1912.00965/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Dec 2019.

## Built on by (most cited first, 8 of 8)

- [Training Over-parameterized Models with Non-decomposable Objectives](https://mlsanity.com/paper/2107.04641/) (2021)
- [Implicit Rate-Constrained Optimization of Non-decomposable Objectives](https://mlsanity.com/paper/2107.10960/) (2021)
- [Principled Algorithms for Optimizing Generalized Metrics in Binary Classification](https://mlsanity.com/paper/2512.23133/) (2025)
- [Optimal Binary Classification Beyond Accuracy](https://mlsanity.com/paper/2107.01777/) (2021)
- [Exact Reformulation and Optimization for Direct Metric Optimization in Binary Imbalanced Classification](https://mlsanity.com/paper/2507.15240/) (2025)
- [Aligning Multiclass Neural Network Classifier Criterion with Task Performance Metrics](https://mlsanity.com/paper/2405.20954/) (2024)
- [Partial VOROS: A Cost-aware Performance Metric for Binary Classifiers with Precision and Capacity Constraints](https://mlsanity.com/paper/2510.18520/) (2025)
- [Principled Algorithms for Optimizing Generalized Metrics in Multi-Label Learning](https://mlsanity.com/paper/2605.28767/) (2026)

## Abstract (arXiv)

We propose a method that enables practitioners to conveniently incorporate custom non-decomposable performance metrics into differentiable learning pipelines, notably those based upon neural network architectures. Our approach is based on the recently developed adversarial prediction framework, a distributionally robust approach that optimizes a metric in the worst case given the statistical summary of the empirical distribution. We formulate a marginal distribution technique to reduce the complexity of optimizing the adversarial prediction formulation over a vast range of non-decomposable metrics. We demonstrate how easy it is to write and incorporate complex custom metrics using our provided tool. Finally, we show the effectiveness of our approach various classification tasks on tabular datasets from the UCI repository and benchmark datasets, as well as image classification tasks. The code for our proposed method is available at https://github.com/rizalzaf/AdversarialPrediction.jl.