Plastic self-adaptation, nonlinear recurrent dynamics and multi-scale memory are desired features in hardware implementations of neural networks, because they enable them to learn, adapt and process information similarly to the way biological brains do. In this work, we experimentally demonstrate these properties occurring in arrays of photonic neurons. Importantly, this is realised autonomously in an emergent fashion, without the need for an external controller setting weights and without explicit feedback of a global reward signal. Using a hierarchy of such arrays coupled to a backpropagation-free training algorithm based on simple logistic regression, we are able to achieve a performance of 98.2% on the MNIST task, a popular benchmark task looking at classification of written digits. The plastic nodes consist of silicon photonics microring resonators covered by a patch of phase-change material that implements nonvolatile memory. The system is compact, robust, and straightforward to scale up through the use of multiple wavelengths. Moreover, it constitutes a unique platform to test and efficiently implement biologically plausible learning schemes at a high processing speed.
A native signed incoherent photonic matrix core on thin-film lithium niobate supporting in situ backpropagation
East China Normal University(华东师范大学);Hefei National Laboratory(合肥国家实验室);Shanghai Research Center for Quantum Sciences(上海量子科学研究中心);Shanxi University(山西大学)
Monadic Pavlovian associative learning in a backpropagation-free photonic network
一个多世纪以前,伊万·巴甫洛夫(Ivan P. Pavlov)在一个经典的实验中展示了狗如何学会将铃声与食物联系起来,从而导致铃声导致唾液分泌。今天,很少发现巴甫洛夫类型的联想学习用于人工智能(AI)应用,即使其他学习概念,特别是人工神经网络(ANN)的反向传播已经蓬勃发展。然而,在“传统”ANN上使用反向传播方法进行训练,特别是以现代深度神经网络(DNN)的形式,是计算和能量密集型的。在这里,我们通过实验证明了一种使用单个(或一元)关联硬件元素的反向传播自由学习形式。我们实现了这一集成的光子平台上使用相变材料与片上级联定向耦合器相结合。然后,我们使用我们的一元巴甫洛夫光子硬件开发了一个放大的电路网络,该硬件提供了一个基于单元素关联的独特机器学习框架,重要的是,使用无反向传播架构来解决一般的学习任务。我们的方法减少了传统神经网络方法中学习所带来的计算负担,从而提高了速度,同时也为我们的光子实现提供了更高的带宽。
Emergent learning: neuromorphic photonic computing with accelerated training
Italian Institute of Technology(意大利理工学院);University Sapienza(罗马第一大学);CINECA(CINECA计算中心);Zhejiang University(浙江大学);Institute of Nanotechnology of the National Research Council of Italy, CNR-NANOTEC, Rome Unit(意大利国家研究委员会纳米技术研究所,CNR-NANOTEC罗马分部)
Silicon Photonic Architecture for Training Deep Neural Networks with Direct Feedback Alignment
Queen’s University(女王大学);University of British Columbia(不列颠哥伦比亚大学);George Washington University(乔治华盛顿大学);Princeton University(普林斯顿大学);Vector Institute(向量研究所)
Dual adaptive training of photonic neural networks
Tsinghua University(清华大学);Shanghai Jiao Tong University(上海交通大学);Beijing National Research Center for Information Science and Technology(北京信息科学与技术国家研究中心)
Toward all-optical unsupervised Hebbian learning in deep photonic neuromorphic networks
Department of Electrical and Computer Engineering, North Carolina State University(北卡罗来纳州立大学电气与计算机工程系);Department of Electrical and Computer Engineering, The University of Texas at Dallas(德克萨斯大学达拉斯分校电气与计算机工程系);Department of Materials Science and Engineering, North Carolina State University(北卡罗来纳州立大学材料科学与工程系);Department of Physics, North Carolina State University(北卡罗来纳州立大学物理系)