Research map / Architectures and efficiency / RNNs and normalization
Simultaneous embedding of multiple attractor manifolds in a recurrent neural network using constrained gradient optimization
Haggai Agmon, Yoram Burak
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 |
|---|---|---|---|
| Embed ten maps instead of one in the connectivity (ablation) | Ring attractor network · Bump drift simulations, L=1 vs L=10 | bump position drift | Hurt |
| Increase number of embedded maps L (ablation) | Unmodified ring attractor network · Drift and energy landscape roughness | drift magnitude | Hurt |
| Add weight modifications from idealized bump energy flattening | Ring attractor network with L maps · L = 10 maps | energy landscape flatness and drift | Improved |
| Add weight modifications from five constrained gradient optimization iterations instead of idealized-bump modifications (ablation) | Ring attractor network · Network stability versus L | drift and firing rate change | Improved |
| Add weight modifications | Ring attractor network · Bump score versus L | bump score / capacity: kink at L = 20 (N = 600) | No clear effect |
| Use energy-landscape flattening via weight modifications that cover all maps simultaneously | Multi-map attractor network · Drift reduction | drift | Improved |
Closest papers by meaning
- Back to the Continuous Attractor (NeurIPS 2024 · similarity 0.90)
- Dynamical phases of short-term memory mechanisms in RNNs (ICML 2025 · similarity 0.88)
- Unconditional stability of a recurrent neural circuit implementing divisive normalization (NeurIPS 2024 · similarity 0.86)
- Partial observation can induce mechanistic mismatches in data-constrained models of neural dynamics (NeurIPS 2024 · similarity 0.86)
- Learning Reveals Invisible Structure in Low-Rank RNNs (NeurIPS 2026 · similarity 0.86)
- Scaling Laws and Tradeoffs in Recurrent Networks of Expressive Neurons (NeurIPS 2026 · similarity 0.86)
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.
