Research map / Probabilistic and scientific modeling / Dynamical systems
Learning Attractor Dynamics for Generative Memory
Yan Wu, Gregory Wayne, Karol Gregor, Timothy Lillicrap
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 Dynamic Kanerva Machine instead of Kanerva Machine | Kanerva Machine · Omniglot test | conditional ELBO: 77.2 vs 68.3 | Hurt |
| Use parameter-free dynamic addressing rather than parametrized addressing | Dynamic Kanerva Machine · Omniglot episodes longer than training episodes | retrieval error | Improved |
| Run iterative attractor retrieval on inputs corrupted by unseen noise | Dynamic Kanerva Machine · Omniglot and DMLab denoising | image quality | Improved |
| Run iterative attractor retrieval on low-quality prior samples | Dynamic Kanerva Machine · Omniglot and DMLab sampling | sample quality | Improved |
| Add extra iterations that re-address using the updated memory during memory writing (ablation) | Dynamic Kanerva Machine · memory writing | lower-bound tightness | Improved |
Closest papers by meaning
- The balance between feature learning and collapse in generative dynamical systems (NeurIPS 2026 · similarity 0.88)
- Back to the Continuous Attractor (NeurIPS 2024 · similarity 0.87)
- Predictive Attractor Models (NeurIPS 2024 · similarity 0.87)
- A solvable model of learning generative diffusion: theory and insights (NeurIPS 2025 · similarity 0.86)
- Equilibrium Reasoners: Learning Attractors Enables Scalable Reasoning (ICML 2026 · similarity 0.85)
- Generalization at the Edge of Stability (NeurIPS 2026 · similarity 0.85)
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.
