Scholarlink

<p>In-sensor computing (ISC) monolithically integrates photodetection and neural-network inference on the same focal plane, avoiding analog-to-digital conversion and off-chip data movement. Despite its unrivalled latency and energy benefits, ISC remains limited by the fragility of analog weights. Amplitude-dependent drift, random-telegraph noise, and irreversible <i>stuck-at</i> faults can each trigger abrupt accuracy collapse. We introduce a training-free hardening strategy—<i>signed-weight projection</i>—that clips every weight to a symmetric range, enforcing a zero-mean distribution without changing network topology or requiring retraining. First-order perturbation analysis shows that the resulting balanced weights self-cancel the mean shift produced by multiplicative noise, while the magnitude cap limits the worst-case impact of saturated faults. Hardware measurements on a <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(10\times 10\)</EquationSource> </InlineEquation> ISC prototype confirm a full optical-to-electrical inference path in <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\sim 2\,\mu \text {s}\)</EquationSource> </InlineEquation>, underscoring the need for on-array robustness. Device-level simulations on a (784–100–10) classifier further demonstrate graceful accuracy degradation. Noise tolerance and stuck-fault tolerance expand by well over an order of magnitude compared with an unprotected baseline, yet clean-data accuracy is preserved. The method offers a low-cost algorithm device co-design guideline, which can be retro-fitted to existing ISC pipelines and scaled to future high-resolution vision sensors. </p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Signed-weight projection for robust in-sensor computing neural classifiers under amplitude-dependent noise and stuck-at faults

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月9日

查看全文
  • Difei Zhong

摘要

In-sensor computing (ISC) monolithically integrates photodetection and neural-network inference on the same focal plane, avoiding analog-to-digital conversion and off-chip data movement. Despite its unrivalled latency and energy benefits, ISC remains limited by the fragility of analog weights. Amplitude-dependent drift, random-telegraph noise, and irreversible stuck-at faults can each trigger abrupt accuracy collapse. We introduce a training-free hardening strategy—signed-weight projection—that clips every weight to a symmetric range, enforcing a zero-mean distribution without changing network topology or requiring retraining. First-order perturbation analysis shows that the resulting balanced weights self-cancel the mean shift produced by multiplicative noise, while the magnitude cap limits the worst-case impact of saturated faults. Hardware measurements on a \(10\times 10\) ISC prototype confirm a full optical-to-electrical inference path in \(\sim 2\,\mu \text {s}\) , underscoring the need for on-array robustness. Device-level simulations on a (784–100–10) classifier further demonstrate graceful accuracy degradation. Noise tolerance and stuck-fault tolerance expand by well over an order of magnitude compared with an unprotected baseline, yet clean-data accuracy is preserved. The method offers a low-cost algorithm device co-design guideline, which can be retro-fitted to existing ISC pipelines and scaled to future high-resolution vision sensors.