Analog in-memory computing (AIMC) performs computation directly within resistive crossbar arrays, offering an energy-efficient platform to scale large vision and language models. However, non-ideal analog device properties make the training on AIMC devices challenging. In particular, its update asymmetry can induce a systematic drift of weight updates towards a device-specific symmetric point (SP), which typically does not align with the optimum of the training objective. To mitigate this bias, most existing works assume the SP is known and pre-calibrate it to zero before training by setting the reference point as the SP. Nevertheless, calibrating AIMC devices requires costly pulse updates, and residual calibration error can directly degrade training accuracy. In this work, we present the first theoretical characterization of the pulse complexity of SP calibration and the resulting estimation error. We further propose a dynamic SP estimation method that tracks the SP during model training, and establishes its convergence guarantees. In addition, we develop an enhanced variant based on chopping and filtering techniques from digital signal processing. Numerical experiments demonstrate both the efficiency and effectiveness of the proposed method.

翻译：模拟内存计算（AIMC）直接在电阻交叉阵列中执行计算，为扩展大规模视觉和语言模型提供了一个高能效的平台。然而，模拟器件的非理想特性使得在AIMC设备上进行训练具有挑战性。具体而言，其更新不对称性会导致权重更新向器件特定的对称点（SP）发生系统性漂移，而该对称点通常与训练目标的最优点不一致。为减轻这种偏差，现有研究大多假设对称点已知，并在训练前通过将参考点设为对称点来将其预校准至零。然而，校准AIMC设备需要成本高昂的脉冲更新，且残留的校准误差会直接降低训练精度。在本工作中，我们首次从理论上刻画了对称点校准的脉冲复杂度及其导致的估计误差。我们进一步提出了一种动态对称点估计方法，该方法在模型训练过程中追踪对称点，并建立了其收敛性保证。此外，我们基于数字信号处理中的斩波与滤波技术，开发了一种增强型变体。数值实验证明了所提方法的高效性与有效性。