Presentation
VeRA+: Vector-Based Lightweight Compensation for Drift-Resilient RRAM In-Memory Computing
DescriptionRRAM-based in-memory computing (IMC) offers high energy efficiency
but suffers from conductance drift that severely degrades long-term
accuracy. Existing approaches including retraining, noise-aware
training, and Batch Normalization-based calibration either require RRAM rewriting,
demand large storage overhead, or rely on online correction. We
propose VeRA+, a lightweight drift compensation framework that
reuses shared projection matrices and introduces only two compact
drift-specific vectors per drift level. A drift-aware scheduling
algorithm offline-trains a small set of VeRA+ parameters and
selects the appropriate set over time without any on-chip
retraining or data replay. VeRA+ preserves up to 99.77% of the drift-free accuracy after ten years of simulated drift and reduces storage overhead by more than three orders of magnitude compared with BN-based calibration. To validate VeRA+ under realistic device behavior, we extract one-week drift statistics from measurements on our fabricated 1T1R RRAM devices and use them to simulate realistic drifted weights. Under these measured drift conditions,
VeRA+ achieves accuracy close to the drift-free baseline, providing an
efficient and practical solution for long-term drift resilience in RRAM-IMC.
but suffers from conductance drift that severely degrades long-term
accuracy. Existing approaches including retraining, noise-aware
training, and Batch Normalization-based calibration either require RRAM rewriting,
demand large storage overhead, or rely on online correction. We
propose VeRA+, a lightweight drift compensation framework that
reuses shared projection matrices and introduces only two compact
drift-specific vectors per drift level. A drift-aware scheduling
algorithm offline-trains a small set of VeRA+ parameters and
selects the appropriate set over time without any on-chip
retraining or data replay. VeRA+ preserves up to 99.77% of the drift-free accuracy after ten years of simulated drift and reduces storage overhead by more than three orders of magnitude compared with BN-based calibration. To validate VeRA+ under realistic device behavior, we extract one-week drift statistics from measurements on our fabricated 1T1R RRAM devices and use them to simulate realistic drifted weights. Under these measured drift conditions,
VeRA+ achieves accuracy close to the drift-free baseline, providing an
efficient and practical solution for long-term drift resilience in RRAM-IMC.
Event Type
Research Manuscript
TimeWednesday, July 291:30pm - 1:42pm PDT
LocationMtg Room 202C
Similar Presentations
