方位辨别知觉学习的眼传递及空间频率调谐特性

为了研究方位辨别知觉学习的神经机制,采用心理物理的方法,研究了方位辨别知觉学习的空间频率调谐与眼传递特性.通过对9个被试在固定空间频率（周期/度）进行方位辨别的训练,并于训练前后在多个空间频率（0.5, 1, 2, 4, 8, 16和32周期/度）测量方位辨别的阈值,发现被试在训练空间频率的方位辨别阈值平均下降了7.01 dB,并且这种学习效应可显著传递到非训练眼.学习效应可分为两个部分：一部分可传递到其他空间频率（4.10 dB）,另外一部分则特异于训练空间频率（2.95 dB）.这些结果说明方位辨别学习可能发生在双眼信息汇聚之后的视皮层,且可能有多种机制参与其中.     

Practising orientation identification improves orientation coding in V1 neurons

The adult brain shows remarkable plasticity, as demonstrated by the improvement in fine sensorial discriminations after intensive practice. The behavioural aspects of such perceptual learning are well documented, especially in the visual system. Specificity for stimulus attributes clearly implicates an early cortical site, where receptive fields retain fine selectivity for these attributes; however, the neuronal correlates of a simple visual discrimination task remained unidentified. Here we report electrophysiological correlates in the primary visual cortex (V1) of monkeys for learning orientation identification. We link the behavioural improvement in this type of learning to an improved neuronal performance of trained compared to naive neurons. Improved long-term neuronal performance resulted from changes in the characteristics of orientation tuning of individual neurons. More particularly, the slope of the orientation tuning curve that was measured at the trained orientation increased only for the subgroup of trained neurons most likely to code the orientation identified by the monkey. No modifications of the tuning curve were observed for orientations for which the monkey had not been trained. Thus training induces a specific and efficient increase in neuronal sensitivity in V1.     

Signal but not noise changes with perceptual learning

Perceptual discrimination improves with practice. This 'perceptual learning' is often specific to the stimuli presented during training, indicating that practice may alter the response characteristics of cortical sensory neurons. Although much is known about how learning modifies cortical circuits, it remains unclear how these changes relate to behaviour. Different theories assume that practice improves discrimination by enhancing the signal, diminishing internal noise or both. Here, to distinguish among these alternatives, we fashioned sets of faces and textures whose signal strength could be varied, and we trained observers to identify these patterns embedded in noise. Performance increased by up to 400% across several sessions over several days. Comparisons of human performance to that of an ideal discriminator showed that learning increased the efficiency with which observers encoded task-relevant information. Observer response consistency, measured by a double-pass technique in which identical stimuli are shown twice in each experimental session, did not change during training, showing that learning had no effect on internal noise. These results indicate that perceptual learning may enhance signal strength, and provide important constraints for theories of learning.     

Involvement of visual cortex in tactile discrimination of orientation.

The primary sense modalities (vision, touch and so on) are generally thought of as distinct. However, visual imagery is implicated in the normal tactile perception of some object properties, such as orientation, shape and size. Furthermore, certain tactile tasks, such as discrimination of grating orientation and object recognition, are associated with activity in areas of visual cortex. Here we show that disrupting function of the occipital cortex using focal transcranial magnetic stimulation (TMS) interferes with the tactile discrimination of grating orientation. The specificity of this effect is illustrated by its time course and spatial restriction over the scalp, and by the failure of occipital TMS to affect either detection of an electrical stimulus applied to the fingerpad or tactile discrimination of grating texture. In contrast, TMS over the somatosensory cortex blocked discrimination of grating texture as well as orientation. We also report that, during tactile discrimination of grating orientation, an evoked potential is recorded over posterior scalp regions with a latency corresponding to the peak of the TMS interference effect (about 180 ms). The findings indicate that visual cortex is closely involved in tactile discrimination of orientation. To our knowledge, this is the first demonstration that visual cortical processing is necessary for normal tactile perception.     

Activation of specific interneurons improves V1 feature selectivity and visual perception

Inhibitory interneurons are essential components of the neural circuits underlying various brain functions. In the neocortex, a large diversity of GABA (γ-aminobutyric acid) interneurons has been identified on the basis of their morphology, molecular markers, biophysical properties and innervation pattern. However, how the activity of each subtype of interneurons contributes to sensory processing remains unclear. Here we show that optogenetic activation of parvalbumin-positive (PV+) interneurons in the mouse primary visual cortex (V1) sharpens neuronal feature selectivity and improves perceptual discrimination. Using multichannel recording with silicon probes and channelrhodopsin-2 (ChR2)-mediated optical activation, we found that increased spiking of PV+ interneurons markedly sharpened orientation tuning and enhanced direction selectivity of nearby neurons. These effects were caused by the activation of inhibitory neurons rather than a decreased spiking of excitatory neurons, as archaerhodopsin-3 (Arch)-mediated optical silencing of calcium/calmodulin-dependent protein kinase IIα (CAMKIIα)-positive excitatory neurons caused no significant change in V1 stimulus selectivity. Moreover, the improved selectivity specifically required PV+ neuron activation, as activating somatostatin or vasointestinal peptide interneurons had no significant effect. Notably, PV+ neuron activation in awake mice caused a significant improvement in their orientation discrimination, mirroring the sharpened V1 orientation tuning. Together, these results provide the first demonstration that visual coding and perception can be improved by increased spiking of a specific subtype of cortical inhibitory interneurons.     

视知觉训练与初级视皮层神经元信息及其表征容量

通过分析猕猴初级视皮层（V1）神经元的群体活动在轮廓线检测训练过程中的变化发现：知觉训练可以降低V1神经元响应在不同试次之间的变异性，促使V1神经元对相同刺激的响应更加稳定；训练还可降低V1神经元群体中神经元活动之间的相关性，使V1神经元之间的活动更加独立；增加了神经活动的维度，提高了神经元表征信息的容量，减少了冗余的神经活动，进而提升了感知能力.     

自适应学习速率法在变压器故障诊断中的应用

为了提高电力变压器故障诊断的准确率,针对油中溶解气体分析,提出了一种基于误差自动调节修正因子的自适应学习速率法,使神经网络通过自身的误差变化过程自动调整学习速率修正因子,保证网络总是以最大的可接受学习速率进行训练,从而提高网络收敛速度。针对电力变压器故障气体及故障类型的特点,建立了电力变压器故障诊断BP(Back-Propagation)网络模型,应用该算法和原算法对该故障诊断网络模型进行训练。仿真结果表明,该算法的训练次数减少了35.4%,收敛速度提高了44.9%,有效地改善了网络模型的性能。将该算法应用于电力变压器故障诊断,能较为精确地判断出电力变压器的故障类型,故障诊断准确率达90.8%。     

激素敏感脂酶(HSL)基因敲除小鼠认知行为的研究

目的:评估HSL基因敲除小鼠的学习记忆功能.方法:采用经典的Morris水迷宫实验,测试三种基因型小鼠的认知行为变化.结果:三种基因型小鼠的潜伏期几乎都逐渐缩短,雄性正常小鼠的潜伏期在1～2天,明显地短于HSL敲除小鼠(P<0.01);雌性正常小鼠的潜伏期在第3～5天,明显地短于HSL敲除小鼠(P<0.01),随后几天都无显著差异.第8天撤去平台,小鼠跨越原平台位置的次数无显著差异.结论:HSL敲除的小鼠都存在明显的认知水平功能障碍,雄性小鼠在前期差异显著,雌性小鼠在中期差异显著,但给予适当的学习训练后都能达到正常小鼠的认知水平.     

Self-organizing neural network that discovers surfaces in random-dot stereograms.

The standard form of back-propagation learning is implausible as a model of perceptual learning because it requires an external teacher to specify the desired output of the network. We show how the external teacher can be replaced by internally derived teaching signals. These signals are generated by using the assumption that different parts of the perceptual input have common causes in the external world. Small modules that look at separate but related parts of the perceptual input discover these common causes by striving to produce outputs that agree with each other. The modules may look at different modalities (such as vision and touch), or the same modality at different times (for example, the consecutive two-dimensional views of a rotating three-dimensional object), or even spatially adjacent parts of the same image. Our simulations show that when our learning procedure is applied to adjacent patches of two-dimensional images, it allows a neural network that has no prior knowledge of the third dimension to discovery depth in random dot stereograms of curved surfaces.     

健身气功·易筋经锻炼对老年人身体机能和血脂的影响

目的:探讨健身气功.易筋经锻炼对老年人身体机能和血脂的影响。方法:以70名老年女性为对象,随机分为练功组(n=35)和对照组(n=35),练功组进行每周5次,每次60分钟,持续3个月的锻炼,对照组不参加任何集体或个人的有计划锻炼。分别于实验前和3个月练功后对实验对象的血压、肺活量、坐位体前屈、握力、背力、反应时、闭眼单腿站立、TC、TG、HDL、LDL测试。结果:练功3个月后,练功组舒张压显著降低(p<0.05),坐位体前屈明显增加(p<0.01)、背力增加(p<0.05)、反应时减少(p<0.05)、闭眼单腿站立时间明显增加(p<0.01),TG降低(p<0.05)、HDL升高(p<0.05)、LDL显著降低(p<0.01);对照组在实验前后所测指标均未发生明显变化(p>0.05)。结论:3个月的健身气功.易筋经锻炼,对老年女性的身体机能和血脂产生了积极的健身效果。     

观众效应对网球训练的影响

观众效应是运动心理学关注的一个热点问题．以浙江师范大学行知学院体育专业的36名从未接触过网球的新生为被试,通过3个月的网球教学训练,检验观众效应对其训练阶段的影响．结果表明：1）正手原地击球、反手原地击球和正反手移动击球的阶段主效应显著;发球阶段主效应接近显著,即后期阶段的训练效果好于早期阶段．2）正反手移动击球训练的“有无观众”与“训练阶段”的交互作用显著,F=34．596,P〈0．01;3）“有无观众”与“训练阶段”的交互作用可以作为检测网球动作技能掌握熟练程度的指标．因此,在网球训练特别是动作技能基本熟练掌握后的训练中应该注重观众效应对训练效果的影响,充分利用观众效应的促进作用,避免观众效应的消极作用．     

基于BP神经网络的光栅莫尔条纹细分方法

一种基于BP神经网络的光栅莫尔条纹细分方法可以提高光栅莫尔条纹的细分精度。在1个莫尔条纹信号周期内采集多个样本点进行差值细分,利用3层BP神经网络,将获得的差值作为网络输入,该样本点在1个栅距内的微位移量作为目标输出,建立合理的网络模型,与DSP微处理器相结合实现莫尔条纹细分。通过对样本点的学习,莫尔条纹细分精度可以达到0．001微米级。同时对非样本点进行验证,实验结果表明该系统具有可行性。     

学习后额叶皮层突触传递效能的探讨

应用电生理学技术结合行为学方法,探查了大鼠在明暗分辨学习后额叶皮层的突触效能变化,结果表明：多数动物（64．7％）出现突触传递效能的增强变化,少数（11．8％）出现降低变化或不明显变化（23．5％）;并且,学习后检测到的突触效能增强变化还可随行为作业的迅速变更而相应改变,即随行为的消退而恢复,又随行为的再学习而再次出现增强变化,本文并就结果予以讨论。     

提高BP神经网络学习速度的算法研究

本文在深入分析了ＢＰ人工神经网络学习算法的基础上，研究了采用加动量项法、成批训练法及综合法来提高ＢＰ网络的学习速度，取得了显著效果．     

基于 LVQ 过程神经元网络的储层岩性识别

针对基于取心井岩心分析数据和测井过程数据的储层岩性判别问题, 建立了一类学习向量量化过程神经 元网络模型(LVQ-PNN: Learning Vector Quantization Process Neural Network)。 该模型通过增加输出层, 扩展了 自组织过程神经元网络的深度结构; 采用无监督竞争与有教师示教相结合的算法策略, 提高了多维信号特征的 自适应提取和自组织综合能力。 实验证明, 该方法具有较好的岩性特征综合和辨识能力, 岩性识别率达到了 84. 7%。     

人工神经网络预测离子色谱分离条件

 以EDTA为淋洗液,用均匀试验设计法研究了高Cl^-浓度下Cl^-与NO₂^-,NO₃^-的分离条件.将实验条件和实验结果作为BP神经网络的训练集,同时对网络的结构和训练速率进行筛选和优化.经过2567次训练后,网络训练误差达到10^-5.对训练结果的考察表明,人工神经网络以较少的实验数据成功建立了离子色谱分离条件的预测模型,最大相对误差仅为0.85%.     

Combinatorial Optimization Based Analog Circuit Fault Diagnosis with Back Propagation Neural Network

Electronic components＇ reliability has become the key of the complex system mission execution. Analog circuit is an important part of electronic components. Its fault diagnosis is far more challenging than that of digital circuit. Simulations and applications have shown that the methods based on BP neural network are effective in analog circuit fault diagnosis. Aiming at the tolerance of analog circuit,a combinatorial optimization diagnosis scheme was proposed with back propagation（ BP） neural network（ BPNN）.The main contributions of this scheme included two parts：（ 1） the random tolerance samples were added into the nominal training samples to establish new training samples,which were used to train the BP neural network based diagnosis model;（ 2） the initial weights of the BP neural network were optimized by genetic algorithm（ GA） to avoid local minima,and the BP neural network was tuned with Levenberg-Marquardt algorithm（ LMA） in the local solution space to look for the optimum solution or approximate optimal solutions. The experimental results show preliminarily that the scheme substantially improves the whole learning process approximation and generalization ability,and effectively promotes analog circuit fault diagnosis performance based on BPNN.     

基于单词领域特征敏感的多领域神经机器翻译

鉴于现有基于单词的领域特征学习方法在领域识别上的精度较低,为提高领域判别和提供准确的翻译,提出一种单词级别的领域特征敏感学习机制,包括两方面:1)编码器端的上下文特征编码,为了扩展单词级别的领域特征学习范围,引入卷积神经网络,并行提取不同大小窗口的词串作为单词的上下文特征;2)强化的领域特征学习,设计基于多层感知机的领域判别器模块,增强从单词上下文特征中获取更准确领域比例的学习能力,提升单词的领域判别准确率。在多领域UM-Corpus英–汉和OPUS英–法翻译任务中的实验结果显示,所提方法平均BLEU值分别超过强基线模型0.82和1.06,单词的领域判别准确率比基线模型分别提升10.07%和18.06%。对实验结果的进一步分析表明,所提翻译模型性能的提升得益于所提出的单词领域特征敏感的学习机制。     

基于步行时足底压力信息的前交叉韧带断裂辅助诊断方法

为了辨识动态足底压力信息与前交叉韧带断裂的关系, 将步行时的足底压力数据转换成图像, 采用深度学习中的卷积神经网络模型, 在给定足量输入图像与分类结果的情况下, 不断更新神经网络的参数, 建立图像与前交叉韧带断裂的关系。将足底压力测试系统(FootScan^®)采集的数据分为训练集和测试集两个部分。训练集用于调整模型的参数, 帮助模型更好地分析并找到足底压力信息与前交叉韧带断裂的关系; 测试集用于模拟诊断, 对比真实情况, 评估准确性, 并评估其作为临床辅助诊断方法的性能。结果表明, 提出的投票法模型的诊断正确率超过90%, 并且从得到足底压力数据到产生诊断结果, 总耗时仅3秒左右。由此得出, 所提出的基于步行时足底压力信息的深度学习模型, 可以在很短时间内辅助诊断前交叉韧带断裂, 为临床辅助诊断及康复提供参考。     

基于迁移学习和支持向量机的白细胞分类

针对人工镜检分类白细胞准确率和效率低的问题,基于深度学习和机器学习算法,提出了一种基于迁移学习和支持向量机的白细胞分类方法.首先对迁移模型进行微调训练,其次用微调训练后的迁移模型进行特征提取,然后将特征输入至神经网络和支持向量机中进行训练,最后通过神经网络和支持向量机的组合分类器对白细胞进行分类.实验结果表明,白细胞分类准确率由最初微调训练的83.26％,随着迁移模型的优化提升为90.43％,最后通过组合分类器再次提升为93.52％,可以在临床实践中帮助医生提高诊断的准确率和效率.