# RankNEAT: Outperforming Stochastic Gradient Search in Preference Learning Tasks

arXiv: https://arxiv.org/abs/2204.06901 · HTML: https://mlsanity.com/paper/2204.06901/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 52 references that are in the map.
- First posted on arXiv 14 Apr 2022.

## Builds on (most cited first, 8 of 8)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Three Factors Influencing Minima in SGD](https://mlsanity.com/paper/1711.04623/) (2017)
- [Eigen-CAM: Class Activation Map using Principal Components](https://mlsanity.com/paper/2008.00299/) (2020)
- [DeepRank: A New Deep Architecture for Relevance Ranking in Information Retrieval](https://mlsanity.com/paper/1710.05649/) (2017)
- [On the Relationship Between the OpenAI Evolution Strategy and Stochastic Gradient Descent](https://mlsanity.com/paper/1712.06564/) (2017)
- [The Pixels and Sounds of Emotion: General-Purpose Representations of Arousal in Games](https://mlsanity.com/paper/2101.10706/) (2021)
- [From Pixels to Affect: A Study on Games and Player Experience](https://mlsanity.com/paper/1907.02288/) (2019)
- [Towards General Models of Player Experience: A Study Within Genres](https://mlsanity.com/paper/2110.00978/) (2021)

## Built on by (most cited first, 5 of 5)

- [Affective Game Computing: A Survey](https://mlsanity.com/paper/2309.14104/) (2023)
- [Across-Game Engagement Modelling via Few-Shot Learning](https://mlsanity.com/paper/2409.13002/) (2024)
- [Privileged Contrastive Pretraining for Multimodal Affect Modelling](https://mlsanity.com/paper/2508.03729/) (2025)
- [PREFAB: PREFerence-based Affective Modeling for Low-Budget Self-Annotation](https://mlsanity.com/paper/2601.13904/) (2026)
- [LaScA: Language-Conditioned Scalable Modelling of Affective Dynamics](https://mlsanity.com/paper/2604.07193/) (2026)

## Abstract (arXiv)

Stochastic gradient descent (SGD) is a premium optimization method for training neural networks, especially for learning objectively defined labels such as image objects and events. When a neural network is instead faced with subjectively defined labels--such as human demonstrations or annotations--SGD may struggle to explore the deceptive and noisy loss landscapes caused by the inherent bias and subjectivity of humans. While neural networks are often trained via preference learning algorithms in an effort to eliminate such data noise, the de facto training methods rely on gradient descent. Motivated by the lack of empirical studies on the impact of evolutionary search to the training of preference learners, we introduce the RankNEAT algorithm which learns to rank through neuroevolution of augmenting topologies. We test the hypothesis that RankNEAT outperforms traditional gradient-based preference learning within the affective computing domain, in particular predicting annotated player arousal from the game footage of three dissimilar games. RankNEAT yields superior performances compared to the gradient-based preference learner (RankNet) in the majority of experiments since its architecture optimization capacity acts as an efficient feature selection mechanism, thereby, eliminating overfitting. Results suggest that RankNEAT is a viable and highly efficient evolutionary alternative to preference learning.