Research map / Probabilistic and scientific modeling / Dynamical systems
Model Zoos: A Dataset of Diverse Populations of Neural Network Models
Konstantin Schürholt, Diyar Taskiran, Boris Knyazev, Xavier Giró-i-Nieto, Damian Borth
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Vary hyperparameters instead of varying only the seed | CNN (s) model zoos · MNIST | accuracy diversity (standard deviation): 0.9 -> 30.7 | Improved |
| Fix rather than randomly vary the seeds while varying hyperparameters | Hyp-10 model zoos · MNIST | accuracy distribution: 79.9 (30.7) -> 80.3 (30.3) | No clear effect |
| Use layer-wise weight statistics instead of raw weights as prediction features | Linear model property prediction · MNIST, CNN (s), Seed | accuracy R² (%): 92.3 -> 98.7 | Improved |
| Use layer-wise weight statistics instead of raw weights as prediction features | Linear model property prediction · MNIST, CNN (s), Hyp-10-r | accuracy R² (%): -11.2 -> 69.4 | Improved |
| Use layer-wise weight statistics instead of raw weights as prediction features | Linear model property prediction · CIFAR10, CNN (s), Seed | accuracy R² (%): 75.3 -> 96.0 | Improved |
| Use fixed rather than varied seeds when varying hyperparameters | CNN (s) model zoos · MNIST | accuracy prediction R² (%) from raw weights: -11.2 -> 66.5 | Improved |
Closest papers by meaning
- Explore and Exploit the Diverse Knowledge in Model Zoo for Domain Generalization (ICML 2023 · similarity 0.87)
- Zoo-Tuning: Adaptive Transfer from A Zoo of Models (ICML 2021 · similarity 0.86)
- Hyper-Representations as Generative Models: Sampling Unseen Neural Network Weights (NeurIPS 2022 · similarity 0.86)
- Foundation Models Secretly Understand Neural Network Weights: Enhancing Hypernetwork Architectures with Foundation Models (ICLR 2025 · similarity 0.84)
- Understanding the Differences in Foundation Models: Attention, State Space Models, and Recurrent Neural Networks (NeurIPS 2024 · similarity 0.84)
- BLEND: Behavior-guided Neural Population Dynamics Modeling via Privileged Knowledge Distillation (ICLR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
