# Scaling Laws for Associative Memories

Preprint · Vivien Cabannes, Elvis Dohmatob, Alberto Bietti

- Archive id: arxiv:2310.02984
- Published: 2023-10-04
- arXiv: https://arxiv.org/abs/2310.02984
- Citations: 0
- Source URL: https://arxiv.org/abs/2310.02984
- PDF: https://arxiv.org/pdf/2310.02984
- Provenance: arxiv

## Abstract

Learning arguably involves the discovery and memorization of abstract rules. The aim of this paper is to study associative memory mechanisms. Our model is based on high-dimensional matrices consisting of outer products of embeddings, which relates to the inner layers of transformer language models. We derive precise scaling laws with respect to sample size and parameter size, and discuss the statistical efficiency of different estimators, including optimization-based algorithms. We provide extensive numerical experiments to validate and interpret theoretical results, including fine-grained visualizations of the stored memory associations.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

Topics: [Scaling Laws & Compute-Optimal Training](https://space-kitty.com/topics/scaling-laws)

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.