Reference library
Meta-Learning through Hebbian Plasticity in Random Networks
Paper summary
Searches for synapse-specific Hebbian learning rules rather than fixed final weights. Randomly initialized networks change during execution and are evaluated on reinforcement-learning tasks, including high-dimensional sensory and locomotion settings.
Relevance to the project
Directly relevant to lifetime adaptation and inherited updater specificity. The paper evolves per-connection coefficients; the project's shared fixed type-level mechanisms pursue a different reuse structure.
Do not describe the inherited coefficients as a single globally shared rule.
Source record
- Reference ID
- neurips-ee23e7ad9b473ad072d57aaa9b2a5222
- Archived version
- NeurIPS 2020 published proceedings PDF
- Summary provenance
- reference-library/papers/neurips-ee23e7ad9b473ad072d57aaa9b2a5222/metadata.json · summary
- Review status
- initial orientation summary; not a critical full-paper review
- Reading scope
- Selected abstract/opening and available closing discussion/conclusion passages in the preserved text extract; broader project notes provide context. No reproduction of experiments, exhaustive methods review, or equation/figure audit.