[论文解读] Neuro-Inspired Visual Pattern Recognition via Biological Reservoir Computing

Luca Ciampi, Ludovico Iannello+5 more|arXiv (Cornell University)|Feb 5, 2026

Neural Networks and Reservoir Computing被引用 0

一句话总结

论文演示了一种生物 reservoir 计算系统，其中在高密度多电极阵列（HD-MEA）上的体外皮层网络作为静态视觉模式识别的储备，与简单点刺激到类似 MNIST 的数字的线性读出进行了评估。

ABSTRACT

In this paper, we present a neuro-inspired approach to reservoir computing (RC) in which a network of in vitro cultured cortical neurons serves as the physical reservoir. Rather than relying on artificial recurrent models to approximate neural dynamics, our biological reservoir computing (BRC) system leverages the spontaneous and stimulus-evoked activity of living neural circuits as its computational substrate. A high-density multi-electrode array (HD-MEA) provides simultaneous stimulation and readout across hundreds of channels: input patterns are delivered through selected electrodes, while the remaining ones capture the resulting high-dimensional neural responses, yielding a biologically grounded feature representation. A linear readout layer (single-layer perceptron) is then trained to classify these reservoir states, enabling the living neural network to perform static visual pattern-recognition tasks within a computer-vision framework. We evaluate the system across a sequence of tasks of increasing difficulty, ranging from pointwise stimuli to oriented bars, clock-digit-like shapes, and handwritten digits from the MNIST dataset. Despite the inherent variability of biological neural responses-arising from noise, spontaneous activity, and inter-session differences-the system consistently generates high-dimensional representations that support accurate classification. These results demonstrate that in vitro cortical networks can function as effective reservoirs for static visual pattern recognition, opening new avenues for integrating living neural substrates into neuromorphic computing frameworks. More broadly, this work contributes to the effort to incorporate biological principles into machine learning and supports the goals of neuro-inspired vision by illustrating how living neural systems can inform the design of efficient and biologically grounded computational models.

研究动机与目标

• 引入一个以培养的皮层网络并与高密度多电极阵列（HD-MEA）接口的生物学基础储备计算范式。
• 校准刺激协议和预处理，以在各会话中诱发稳健、可重复的储备反应。
• 在 progressively 复杂的任务中评估静态视觉模式识别，包括 MNIST 派生输入。

提出的方法

• 使用 HD-MEA 将来自输入图像映射到电极子集的空间电刺激模式进行传递。
• 在非刺激电极上记录诱发活动并转换为 4096‑D 储备状态向量。
• 在两阶段预处理流程中应用伪迹去除和尖峰检测。
• 在多次试验的标记储备状态上，用 SGD 训练单层感知机读出（readout）。
• 在匹配条件下将生物储备的性能与人工储备进行比较。

Figure 1: Schematic overview of our biological reservoir computing (BRC) framework. We introduce a system that encodes spatial stimulation patterns into high‑dimensional neural representations generated by a living cortical network, enabling biologically grounded visual pattern recognition. (1) Inpu

实验结果

研究问题

• RQ1 可以在体外培养的活体皮层网络作为高维储备用于静态视觉模式识别吗？
• RQ2 在逐步复杂的视觉任务中，生物储备相对于匹配的人工储备的表现如何？
• RQ3 刺激协议、记录预处理和会话变异性对下游分类有何影响？
• RQ4 能否使用生物基底对真实世界输入（MNIST 派生模式）实现可靠分类？

主要发现

• 点刺激在 BRC 下实现约 98% 的准确率（均值±SD 98% ± 2%），在此设置下优于人工储备。
• 定向条纹模式在 BRC 下实现约 92% 的准确率（均值±SD 92% ± 6%），当模式变得更重叠时仍具有竞争力。
• 对于时钟-数字样模式，BRC 显示出在映射到电极子集时对数字的显著判别能力（具体数字在摘录中未给出）。
• 在更复杂的任务中，人工储备有时超越 BRC，揭示生物变异性与计算丰富性之间的权衡。
• 研究验证了体外皮层网络可以作为静态模式识别的有效储备，并为将生物基质整合入神经形态框架提供了依据。

Figure 2: Schematic representation of a neuronal culture interfaced with an HD-MEA platform. Each square corresponds to an individual electrode on the HD-MEA. Input patterns are encoded by mapping their elements to specific electrode positions. The remaining electrodes record the evoked spiking acti