An open dataset of neural networks for hypernetwork research
Not on the map. Why?
Abstract (arXiv)
Despite the transformative potential of AI, the concept of neural networks that can produce other neural networks by generating model weights (hypernetworks) has been largely understudied. One of the possible reasons is the lack of available research resources that can be used for the purpose of hypernetwork research. Here we describe a dataset of neural networks, designed for the purpose of hypernetworks research. The dataset includes 104 LeNet-5 neural networks trained for binary image classification separated into 10 classes, such that each class contains 1,000 different neural networks that can identify a certain ImageNette V2 class from all other classes. A computing cluster of over 104 cores was used to generate the dataset. Basic classification results show that the neural networks can be classified with accuracy of 72.0%, indicating that the differences between the neural networks can be identified by supervised machine learning algorithms. The ultimate purpose of the dataset is to enable hypernetworks research. The dataset and the code that generates it are open and accessible to the public.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 17 of its 40 references that are in the map. First posted on arXiv 15 Jul 2025.
Builds on
17 of 40 references.
References · 17
- 2014Microsoft COCO: Common Objects in Contextcited by 17,608
- 2017MoleculeNet: A Benchmark for Molecular Machine Learningcited by 1,051
- 2018Graph HyperNetworks for Neural Architecture Searchcited by 132
- 2020Predicting Neural Network Accuracy from Weightscited by 110
- 2023Equivariant Architectures for Learning in Deep Weight Spacescited by 107
- 2023A Brief Review of Hypernetworks in Deep Learningcited by 102
- 2021Parameter Prediction for Unseen Deep Architecturescited by 89
- 2020Classifying the classifier: dissecting the weight space of neural networkscited by 65
- 2022Hyper-Representations as Generative Models: Sampling Unseen Neural Network Weightscited by 64
- 2021Scientific Machine Learning Benchmarkscited by 60
- 2020Continual Model-Based Reinforcement Learning with Hypernetworkscited by 44
- 2022Model Zoos: A Dataset of Diverse Populations of Neural Network Modelscited by 39
5 more
- 2021Recomposing the Reinforcement Learning Building Blocks with Hypernetworkscited by 35
- 2021AlphaEvolve: A Learning Framework to Discover Novel Alphas in Quantitative Investmentcited by 32
- 2025Unreflected Use of Tabular Data Repositories Can Undermine Research Qualitycited by 9
- 2024Hyper-CL: Conditioning Sentence Representations with Hypernetworkscited by 8
- 2024Analysis and prevention of AI-based phishing email attackscited by 6
Built on by
Not cited yet.