深度神经网络和高光谱显微图像的二维材料纳米片识别

近年来,二维材料由于其独特的性质而受到了广泛关注。在制备二维层状晶体的各种方法中,机械剥离法获得的薄层二维材料晶体质量高,适用于基础研究及性能演示。然而用机械剥离法从衬底上获得的材料具有一定的随机性,可能包含了少许相对较厚的部分。实现对这些二维薄层材料有效、快速且智能化的表征有利于促进二维材料性能的进一步研究。提出了一种基于深度学习的表征方法,通过搭建的编解码结构的卷积神经网络语义分割算法,可以根据光学显微镜图像进行分割和快速识别二维材料纳米片。卷积神经网络作为深度学习在图像处理领域中的典型算法,能够对光学显微镜图像中的复杂信息进行特征提取。首先采用机械剥离制备MoS₂纳米片样本,通过光学显微镜采集高光谱图像并对样本进行标记,根据样本的厚度范围标记出不同的区域,对标记后的图像进一步处理,包括图像的颜色校准和剪切操作,得到用于网络训练和测试的数据集。针对光学图像中二维纳米薄片存在的低对比度、碎裂等特点,编码时加入残差结构和金字塔池化模型,有助于特征信息的提取;解码时融合编码路径中提取的浅层特征信息,以提高网络分割精度。实验中采用带权重的交叉熵损失函数解决类别数量不平衡问题和采用数据增强扩大数据集。对训练后的网络测试结果表明,模型像素精度为97.38%,平均像素精度为90.38%,均交并比为75.86%。之后通过迁移学习成功地对剥离的单层和双层石墨烯纳米片样本进行了识别,均交并比达到了81.63%,表明该方法具有普适性。通过MoS₂和石墨烯纳米片的识别演示,实现了深度学习在二维材料的光学显微镜图像中的成功应用。该方法有望在更多的二维材料上得到扩展并突破自动动态处理光学显微镜图像的问题,同时为其他纳米材料的高光谱图像处理提供参考。     

大规模光电混合神经网络中的可编程拓扑重构

提出了一种在大规模光电混合神经网络系统实现可编程拓扑重构的有效方法，文中介绍了这种方法的原理，运用技术以及采用这种方法在已有的ＮＰ１０２４光学／数字神经网络处理器中运行单层反馈型和多层前馈型等不同网络拓扑结构的实验结果。     

人工神经网络在配煤过程状态建模中的应用研究

本文详细介绍了人工神经网络应用于状态建模的方法．对神经网络应用中的一些难点提出了切实可行且有效的解决措施，并举例作了应用示范．同时还介绍了神经网络方法应用于优化动力配煤的情况，并就神经网络方法在优化动力配煤中的进一步应用作了展望．     

用亚图参数与回归技术估计和预测烷烃的核磁共振碳谱

系统研究了分子建模在波谱分析中的应用.采用多元线性回归算法(MLR)估计和预测了60余种烷烃的碳谱化学位移.烷烃中碳原子由十余种对应于所谓根亚树的相嵌频率描述子所决定.这些描述子等于由2~5个碳原子组成的更小结构骨架组成.说明了所用描述子作为很有用的工具可适当地描述烷烃中碳所处微观环境.同时还比较了与神经网络的计算结果.     

基于条件生成式对抗网络的油藏单井产量预测模型

为克服机器学习方法在油藏单井产量预测中的过拟合问题,提高油田生产中的产量预测精度,提出一种基于条件生成式对抗网络(CGAN)的油藏单井产量预测模型。该模型使用长短期记忆、全连接等基础神经网络,构建生成和判别网络模型。生成网络模型以产量影响因素为条件输入,生成预测产量数据,利用对数损失函数评价预测数据与真实数据之间的偏差,通过条件生成式对抗网络的博弈训练,并结合贝叶斯超参数优化算法,优化模型结构,综合提高模型的泛化能力。基于Eclipse数值模拟软件建立同一井网条件下不同地质和生产条件下的油藏单井产量数据库,以地质与生产条件等产量影响因素作为模型的条件输入,进行油藏单井产量预测。结果表明：与全连接神经网络(FCNN)、随机森林(RF)以及长短期记忆神经网络(LSTM)模型的预测结果相比,CGAN模型在测试集上的平均绝对百分比误差分别提升了2.59%、 0.81%以及1.72%,并且过拟合比最小(1.027)。说明CGAN降低了机器学习产量预测模型的过拟合程度,提高了模型的泛化能力与预测精度,验证了所提算法的优越性,对指导油田高效开发和保障我国能源战略安全具有重要意义。     

基于遗传模拟退火算法的室内照明节能控制方法研究北大核心CSCD

目前的节能照明控制算法仍有陷入局部最优的问题。为了寻求全局最优解,提高室内照明的节能效果,设计一种遗传模拟退火算法对照明系统的控制参数进行优化求解。该算法通过在遗传操作后对优秀个体进行模拟退火处理,增强了算法的局部搜索能力。根据迭代的次数和种群的适应度对遗传概率进行自适应调节,使得算法在前期丰富种群多样性,避免算法“早熟”。提出基于人工神经网络的照度模型来计算室内照度分布,对照明舒适度进行评估,为构造优化算法的适应函数提供了依据。通过仿真实验,在本文介绍的照明场景应用遗传模拟退火算法,并与传统粒子群算法和遗传算法进行比较,其照明节能性能分别高出5.30%和13.61%。     

Identify real gravitational wave events in the LIGO-Virgo catalog GWTC-1 and GWTC-2 with convolutional neural network

In recent years, machine learning models have been introduced into the field of gravitational wave (GW) data processing. In this paper, we apply the convolutional neural network (CNN) to LIGO O1, O2, O3a data analysis to search the released 41 GW events which are emitted from binary black hole (BBH) mergers (here we exclude the events from binary neutron star (BNS) mergers, and the events that are not detected simultaneously by Hanford (H) and Livingston (L) detectors), and use time sliding method to reduce the false alarm rate (FAR). According to the results, the 41 confirmed GW events of BBH mergers can be classified successfully by our CNN model. Furthermore, through restricting the number of consecutive prewarning from sequential samples intercepted continuously in LIGO O2 real time-series and vetoing the coincidences of noise from H and L, the FAR is limited to be less than once in 2 months. It is helpful to promote LIGO real time data processing.     

A Probabilistic Re-Intepretation of Confidence Scores in Multi-Exit Models

In this paper, we propose a new approach to train a deep neural network with multiple intermediate auxiliary classifiers, branching from it. These ‘multi-exits’ models can be used to reduce the inference time by performing early exit on the intermediate branches, if the confidence of the prediction is higher than a threshold. They rely on the assumption that not all the samples require the same amount of processing to yield a good prediction. In this paper, we propose a way to train jointly all the branches of a multi-exit model without hyper-parameters, by weighting the predictions from each branch with a trained confidence score. Each confidence score is an approximation of the real one produced by the branch, and it is calculated and regularized while training the rest of the model. We evaluate our proposal on a set of image classification benchmarks, using different neural models and early-exit stopping criteria.     

Permutation Entropy-Based Interpretability of Convolutional Neural Network Models for Interictal EEG Discrimination of Subjects with Epileptic Seizures vs. Psychogenic Non-Epileptic Seizures

The differential diagnosis of epileptic seizures (ES) and psychogenic non-epileptic seizures (PNES) may be difficult, due to the lack of distinctive clinical features. The interictal electroencephalographic (EEG) signal may also be normal in patients with ES. Innovative diagnostic tools that exploit non-linear EEG analysis and deep learning (DL) could provide important support to physicians for clinical diagnosis. In this work, 18 patients with new-onset ES (12 males, 6 females) and 18 patients with video-recorded PNES (2 males, 16 females) with normal interictal EEG at visual inspection were enrolled. None of them was taking psychotropic drugs. A convolutional neural network (CNN) scheme using DL classification was designed to classify the two categories of subjects (ES vs. PNES). The proposed architecture performs an EEG time-frequency transformation and a classification step with a CNN. The CNN was able to classify the EEG recordings of subjects with ES vs. subjects with PNES with 94.4% accuracy. CNN provided high performance in the assigned binary classification when compared to standard learning algorithms (multi-layer perceptron, support vector machine, linear discriminant analysis and quadratic discriminant analysis). In order to interpret how the CNN achieved this performance, information theoretical analysis was carried out. Specifically, the permutation entropy (PE) of the feature maps was evaluated and compared in the two classes. The achieved results, although preliminary, encourage the use of these innovative techniques to support neurologists in early diagnoses.