Research map / Probabilistic and scientific modeling / Dynamical systems
Hyperbolic neural population geometry benefits computation
Dennis Wu, Yi-Chun Hung, Braden Yuille, James E Fitzgerald, Han Liu
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 |
|---|---|---|---|
| use Karcher-flow hyperbolic attention instead of Hopfield attention | MNIST classifier · MNIST (d=4) | accuracy (%): 83.70 -> 85.52 | Improved |
| use Karcher-flow attention instead of Hopfield attention | MNIST classifier · MNIST (d=8) | accuracy (%): 92.29 -> 92.42 | Improved |
| use Karcher-flow attention instead of Hopfield attention | MNIST classifier · MNIST (d=32) | accuracy (%): 96.71 -> 96.89 | No clear effect |
| use KFPooling instead of HopfieldPooling | MIL · MIL Tiger | AUC: 83.52 -> 87.34 | Improved |
| use KFPooling instead of HopfieldPooling | MIL · MIL Fox | AUC: 60.54 -> 66.00 | Improved |
| use KFPooling instead of HopfieldPooling | MIL · MIL Elephant | AUC: 91.65 -> 91.20 | No clear effect |
| use KFPooling instead of the other hyperbolic baseline A | MIL · MIL Tiger | AUC: 89.20 -> 87.34 | Hurt |
| use KFPooling instead of the other hyperbolic baseline A | MIL · MIL Elephant | AUC: 93.04 -> 91.20 | Hurt |
| use Karcher-flow model instead of the other hyperbolic baseline B | MNIST classifier · MNIST (d=4) | accuracy (%): 67.35 -> 85.52 | Improved |
Closest papers by meaning
- Comparing noisy neural population dynamics using optimal transport distances (ICLR 2025 · similarity 0.85)
- Trading Place for Space: Increasing Location Resolution Reduces Contextual Capacity in Hippocampal Codes (NeurIPS 2024 · similarity 0.85)
- Binding in hippocampal-entorhinal circuits enables compositionality in cognitive maps (NeurIPS 2024 · similarity 0.84)
- Synaptic Weight Distributions Depend on the Geometry of Plasticity (ICLR 2024 · similarity 0.84)
- Place Cells as Multi-Scale Position Embeddings: Random Walk Transition Kernels for Path Planning (NeurIPS 2025 · similarity 0.84)
- A Multi-Region Brain Model to Elucidate the Role of Hippocampus in Spatially Embedded Decision-Making (ICML 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.
