Bridging Silicon and the Hippocampus: Algebro-Deterministic Memory "VaCoAl" as a Substrate for Vector-HaSH and TEM
The paper discusses a new memory architecture called VaCoAl, which integrates concepts from computational neuroscience and hyperdimensional engineering. It proposes that VaCoAl can serve as a substrate for existing models of memory, specifically Vector-HaSH and the Tolman-Eichenbaum Machine. The authors provide formal correspondences and predictions that bridge these fields, emphasizing the architectural similarities with evolutionary conserved pathways in the brain.
- ▪VaCoAl is an algebro-deterministic hyperdimensional memory architecture built from Galois-field LFSRs.
- ▪The architecture provides a substrate-level alternative to existing memory models, satisfying quasi-orthogonality requirements.
- ▪The study derives testable predictions based on the proposed model and its correspondence with hippocampal computation.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Neural and Evolutionary Computing arXiv:2605.15652 (cs) [Submitted on 15 May 2026] Title:Bridging Silicon and the Hippocampus: Algebro-Deterministic Memory "VaCoAl" as a Substrate for Vector-HaSH and TEM Authors:Hiroyuki Chuma, Kanji Otsuka, Yoichi Sato View a PDF of the paper titled Bridging Silicon and the Hippocampus: Algebro-Deterministic Memory "VaCoAl" as a Substrate for Vector-HaSH and TEM, by Hiroyuki Chuma and 2 other authors View PDF HTML (experimental) Abstract:Vector-HaSH and the Tolman-Eichenbaum Machine (TEM) propose that the hippocampal-entorhinal circuit factorizes content from a prestructured grid-cell scaffold and supports compositional memory via ripple-mediated replay.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
