A Drosophila Whole-Connectome Network Can Learn Human-Designed Cognitive Tasks
We use the publicly released MaleCNS v1.0 connectome, reconstructed from a single adult male Drosophila specimen, as the fixed recurrent topology of an artificial network.
ProofPaper ↗
Key points
- Can a biological wiring diagram serve as a useful computational substrate beyond the behaviors for which it evolved?
- We train separate models for bounded addition and for a controlled grounded relational language task built from a fixed 100-word lexicon.
- In both models, one scalar is learned per anatomical edge.
- Together, these results show that higher-order MaleCNS wiring provides a reusable inductive bias for bounded addition and grounded relational language.
Sources (1)
- [1]A Drosophila Whole-Connectome Network Can Learn Human-Designed Cognitive TasksarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 01:04 PM
We use the publicly released MaleCNS v1.0 connectome, reconstructed from a single adult male Drosophila specimen, as the fixed recurrent topology of an artificial network.
Can a biological wiring diagram serve as a useful computational substrate beyond the behaviors for which it evolved?
Extractive summary: sentences quoted from the sources.