Neuromorphic photonics with electro-absorption modulators
Neuromorphic photonics with electro-absorption modulators
复制标题
DOI：
10.1364/oe.27.005181
复制
发表时间：
2019-02-18
期刊：
影响因子：
3.8
通讯作者：
Sorger, Volker J.
中科院分区：
文献类型：
--
作者：
George, Jonathan K.;Mehrabian, Armin;Sorger, Volker J.
光子神经网络受益于高光通道容量以及光的波动特性，光通过线性光学可作为一种有效的加权机制。通过使用有源集成光子元件引入非线性激活函数，能够单片构建多层神经网络，从而无需因外部转换而产生能量和延迟成本。基于干涉仪的调制器虽然在通信中很常用，但已表明其比基于吸收的调制器需要更大的面积，导致神经网络密度降低。在此，我们针对电光全连接神经网络中的吸收调制器建立了一个包含噪声的模型，并将网络性能与五种吸收调制器内在产生的激活函数进行了比较。我们的结果表明，基于量子阱吸收调制器的电光神经元性能最佳，在一个具有2个隐藏层的前馈光子神经网络中进行MNIST分类时，不包括激光功率，其预测准确率可达96%，每MAC能耗为1.7×10⁻¹²焦耳。（根据美国光学学会开放获取出版协议条款，©2019美国光学学会）
Photonic neural networks benefit from both the high-channel capacity and the wave nature of light acting as an effective weighting mechanism through linear optics. Incorporating a nonlinear activation function by using active integrated photonic components allows neural networks with multiple layers to be built monolithically, eliminating the need for energy and latency costs due to external conversion. Interferometer-based modulators, while popular in communications, have been shown to require more area than absorption-based modulators, resulting in a reduced neural network density. Here, we develop a model for absorption modulators in an electro-optic fully connected neural network, including noise, and compare the network's performance with the activation functions produced intrinsically by five types of absorption modulators. Our results show the quantum well absorption modulator-based electro-optic neuron has the best performance allowing for 96% prediction accuracy with 1.7 x 10(-12) J/MAC excluding laser power when performing MNIST classification in a 2 hidden layer feed-forward photonic neural network. (C) 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement