PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
March 3, 2026Nature Communications1 citationsOpen Access

A hardware-adaptive learning algorithm for superlinear-capacity associative memory on memristor crossbars

CHChengping HeMJMingrui JiangKSKeyi Shan

Key Points

  • The approach achieves threefold higher capacity than a pseudo-inverse baseline despite 50% stuck-at faults, indicating remarkable defect tolerance.
  • With effective capacity scaling proportional to N^1.49 and N^1.74 for binary and continuous patterns respectively, strong performance on correlated data is noted.
  • Analysis employed on an integrated memristor crossbar platform enhances recall efficiency by reducing energy usage by 8.8×, showcasing potential for practical applications.
  • This work may enable robust, efficient recall mechanisms, promoting advanced computing architectures through improved algorithm-hardware designs.

Abstract

The human brain recalls complete patterns from partial cues via associative memory, but Hopfield neural networks emulating this process are inefficient on conventional hardware, and prior memristor-based implementations are vulnerable to device defects and have limited capacity, particularly for continuous patterns. We introduce a hardware-adaptive learning algorithm that incorporates experimentally calibrated device constraints during training and validate it on an integrated memristor crossbar compute-in-memory platform. The approach improves defect tolerance and effective capacity, achieving threefold higher capacity than a pseudo-inverse baseline at 50% stuck-at faults. The same framework extends to scalable multilayer architectures supporting binary and continuous-valued patterns, where we observe superlinear capacity scaling on correlated data (∝N1.49 and ∝N1.74, respectively). Leveraging crossbar parallelism with synchronous updates, the implementation reduces energy by 8.8× and latency by 99.7% for 64-dimensional patterns versus asynchronous schemes. These results provide a practical algorithm-hardware co-design for robust, efficient Hopfield-style associative recall.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

He et al. (2026) studied this question.

synapsesocial.com/papers/69a768babadf0bb9e87e5c0chttps://doi.org/10.1038/s41467-026-69958-0
Ask AI
Helpful
Bookmark
Share
View Full Paper