# A Tunable Loss Function for Robust Classification: Calibration, Landscape, and Generalization

arXiv: https://arxiv.org/abs/1906.02314 · HTML: https://mlsanity.com/paper/1906.02314/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Jun 2019.

## Built on by (most cited first, 23 of 23)

- [On Tilted Losses in Machine Learning: Theory and Applications](https://mlsanity.com/paper/2109.06141/) (2021)
- [An Information-theoretical Approach to Semi-supervised Learning under Covariate-shift](https://mlsanity.com/paper/2202.12123/) (2022)
- [Being Properly Improper](https://mlsanity.com/paper/2106.09920/) (2021)
- [Advancing Supervised Learning with the Wave Loss Function: A Robust and Smooth Approach](https://mlsanity.com/paper/2404.18101/) (2024)
- [Realizing GANs via a Tunable Loss Function](https://mlsanity.com/paper/2106.05232/) (2021)
- [Robust PACm: Training Ensemble Models Under Misspecification and Outliers](https://mlsanity.com/paper/2203.01859/) (2022)
- [α-GAN: Convergence and Estimation Guarantees](https://mlsanity.com/paper/2205.06393/) (2022)
- [Addressing GAN Training Instabilities via Tunable Classification Losses](https://mlsanity.com/paper/2310.18291/) (2023)
- [LegendreTron: Uprising Proper Multiclass Loss Learning](https://mlsanity.com/paper/2301.11695/) (2023)
- [(αD,αG)-GANs: Addressing GAN Training Instabilities via Dual Objectives](https://mlsanity.com/paper/2302.14320/) (2023)
- [Uncertainty Quantification and Data Efficiency in AI: An Information-Theoretic Perspective](https://mlsanity.com/paper/2512.05267/) (2025)
- [Xtreme Margin: A Tunable Loss Function for Binary Classification Problems](https://mlsanity.com/paper/2211.00176/) (2022)
- [Generalized Dual Discriminator GANs](https://mlsanity.com/paper/2507.17684/) (2025)
- [Risk-averse Fair Multi-class Classification](https://mlsanity.com/paper/2509.05771/) (2025)
- [AugLoss: A Robust Augmentation-based Fine Tuning Methodology](https://mlsanity.com/paper/2206.02286/) (2022)
- [Smoothly Giving up: Robustness for Simple Models](https://mlsanity.com/paper/2302.09114/) (2023)
- [An extended asymmetric sigmoid with Perceptron (SIGTRON) for imbalanced linear classification](https://mlsanity.com/paper/2312.16043/) (2023)
- [Label Noise Robustness for Domain-Agnostic Fair Corrections via Nearest Neighbors Label Spreading](https://mlsanity.com/paper/2406.09561/) (2024)
- [Ultra-imbalanced classification guided by statistical information](https://mlsanity.com/paper/2409.04101/) (2024)
- [α-GAN by R'enyi Cross Entropy](https://mlsanity.com/paper/2505.14190/) (2025)
- [Risk-Entropic Flow Matching](https://mlsanity.com/paper/2512.03078/) (2025)
- [Minimax Generalized Cross-Entropy](https://mlsanity.com/paper/2603.19874/) (2026)
- [Alpha Diffusion Language Models: Factorization Alone Is Not the Problem](https://mlsanity.com/paper/2609.38066/) (2026)

## Abstract (arXiv)

We introduce a tunable loss function called α-loss, parameterized by α ∈ (0,∞], which interpolates between the exponential loss (α = 1/2), the log-loss (α = 1), and the 0-1 loss (α = ∞), for the machine learning setting of classification. Theoretically, we illustrate a fundamental connection between α-loss and Arimoto conditional entropy, verify the classification-calibration of α-loss in order to demonstrate asymptotic optimality via Rademacher complexity generalization techniques, and build-upon a notion called strictly local quasi-convexity in order to quantitatively characterize the optimization landscape of α-loss. Practically, we perform class imbalance, robustness, and classification experiments on benchmark image datasets using convolutional-neural-networks. Our main practical conclusion is that certain tasks may benefit from tuning α-loss away from log-loss (α = 1), and to this end we provide simple heuristics for the practitioner. In particular, navigating the α hyperparameter can readily provide superior model robustness to label flips (α > 1) and sensitivity to imbalanced classes (α < 1).