False learning function of an acoustic signal classifier

In works on statistical pattern recognition that use learning and examination, results of the learning depend not only on the feature efficiencies, but also on the proportion between the capacity of the decision rule, length of the learning sample, and number of features. It is usually difficult to calculate the recognition errors, which connect these basic quantities for a particular classifier, while the calculations are approximate and do not clearly characterize the results obtained in the process of the study.The purpose of this work is to develop a simple, clear, and efficient technique for the experimental estimation of the expected classification errors of the recognition engine employed in learning. The algorithm produces a sample of random noise segments, which is included in the recognition algorithm instead of the features of real signals. Portions of this uniform sample imitate different classes. The false learning function is produced as a result of a successive increase in the number of random features used in the recognition. The corresponding growth of the probability of recognizing artificial classes in such a false learning depends on the length of the learning sample and on the capacity of the decision rule employed.The main result of this work is the false learning function proposed for any particular classifier. The function is obtained for the same length of the learning sample as that of the one used to recognize real signals. The validity of results obtained in real signals can be estimated by comparing this function with experimental signal recognition probabilities with the same number of features.The simple false learning function is useful to characterize the validity of any experimental results on the statistical signal recognition in acoustics, seismoacoustics, and hydroacoustics; in speech recognition; in medical and industrial diagnostics; in radar; and in other fields.     

A data-efficient self-supervised deep learning model for design and characterization of nanophotonic structures

With its tremendous success in many machine learning and pattern recognition tasks, deep learning, as one type of data-driven models, has also led to many breakthroughs in other disciplines including physics, chemistry and material science. Nevertheless,the supremacy of deep learning over conventional optimization approaches heavily depends on the huge amount of data collected in advance to train the model, which is a common bottleneck of such a data-driven technique. In this work, we present a comprehensive deep learning model for the design and characterization of nanophotonic structures, where a self-supervised learning mechanism is introduced to alleviate the burden of data acquisition. Taking reflective metasurfaces as an example, we demonstrate that the self-supervised deep learning model can effectively utilize randomly generated unlabeled data during training, with the total test loss and prediction accuracy improved by about 15% compared with the fully supervised counterpart.The proposed self-supervised learning scheme provides an efficient solution for deep learning models in some physics-related tasks where labeled data are limited or expensive to collect.     

Perceptual learning and human expertise

We consider perceptual learning: experience-induced changes in the way perceivers extract information. Often neglected in scientific accounts of learning and in instruction, perceptual learning is a fundamental contributor to human expertise and is crucial in domains where humans show remarkable levels of attainment, such as language, chess, music, and mathematics. In Section 2, we give a brief history and discuss the relation of perceptual learning to other forms of learning. We consider in Section 3 several specific phenomena, illustrating the scope and characteristics of perceptual learning, including both discovery and fluency effects. We describe abstract perceptual learning, in which structural relationships are discovered and recognized in novel instances that do not share constituent elements or basic features. In Section 4, we consider primary concepts that have been used to explain and model perceptual learning, including receptive field change, selection, and relational recoding. In Section 5, we consider the scope of perceptual learning, contrasting recent research, focused on simple sensory discriminations, with earlier work that emphasized extraction of invariance from varied instances in more complex tasks. Contrary to some recent views, we argue that perceptual learning should not be confined to changes in early sensory analyzers. Phenomena at various levels, we suggest, can be unified by models that emphasize discovery and selection of relevant information. In a final section, we consider the potential role of perceptual learning in educational settings. Most instruction emphasizes facts and procedures that can be verbalized, whereas expertise depends heavily on implicit pattern recognition and selective extraction skills acquired through perceptual learning. We consider reasons why perceptual learning has not been systematically addressed in traditional instruction, and we describe recent successful efforts to create a technology of perceptual learning in areas such as aviation, mathematics, and medicine. Research in perceptual learning promises to advance scientific accounts of learning, and perceptual learning technology may offer similar promise in improving education.     

多元混沌时间序列的多核极端学习机建模预测

多元混沌时间序列广泛存在于自然、经济、社会、工业等领域. 对多元混沌时间序列进行建模预测有助于人类更好地管理, 控制与决策. 针对多元混沌时间序列的建模预测问题, 本文提出一种基于多核极端学习机的预测方法. 首先对多元混沌时间序列进行相空间重构, 将多元混沌时间序列序列的时间相关性转化为空间相关性. 提出一种结合多核学习算法与核极端学习机模型的多核极端学习机建立相空间中输入输出数据的非线性映射. 多核极端学习机模型结合了多核学习算法的数据融合能力以及核极端学习机的训练简便优势. 基于Lorenz混沌时间序列预测和San Francisco河流月径流量预测的仿真实验表明, 与其他常见混沌时间序列预测方法相比, 本文提出的基于多核极端学习机的多元混沌时间序列预测方法具有更小的预测误差.     

The Large Hadron Collider: A Marvel of Technology,edited by Lyndon Evans

Machine learning algorithms learn a desired input-output relation from examples in order to interpret new inputs. This is important for tasks such as image and speech recognition or strategy optimisation, with growing applications in the IT industry. In the last couple of years, researchers investigated if quantum computing can help to improve classical machine learning algorithms. Ideas range from running computationally costly algorithms or their subroutines efficiently on a quantum computer to the translation of stochastic methods into the language of quantum theory. This contribution gives a systematic overview of the emerging field of quantum machine learning. It presents the approaches as well as technical details in an accessible way, and discusses the potential of a future theory of quantum learning.     

裂缝性介质多尺度深度学习模型

结合人工神经网络建立裂缝介质多尺度深度学习流动模型.基于一套粗网格和一套细网格,通过在粗网格上训练数据,多尺度神经网络能够以较少的自由度训练出准确的神经网络.并在粗网格上通过求解局部流动问题获得多尺度基函数,结合神经网络进一步得到精细网格的解.基于离散裂缝的流动方程可视为多层网络,网络层数依赖于求解时间步数.阐述裂缝介质多尺度机器学习数值计算格式的建立,介绍如何使用多尺度算法构建离散裂缝模型的多尺度基函数,并采用超样本技术进一步提高计算准确性.数值结果表明,多尺度有限元算法与机器学习结合是一种有效的流体流动模拟算法.     

自然循环流动不稳定性的多目标优化极限学习机预测方法

极限学习机是近年来提出的一种前向单隐层神经网络训练算法,具有训练速度快、不会陷入局部最优等优点,但其性能会受到随机选取的输入权值和阈值的影响.针对这一问题,提出一种基于多目标优化的改进极限学习机,将训练误差和输出层权值的均方最小化同时作为优化目标,采用带精英策略的快速非支配排序遗传算法对极限学习机的输入层到隐层的权值和阈值进行优化.将该算法应用于摇摆工况下自然循环系统不规则复合型流量脉动的多步滚动预测,分析了训练误差和输出层权值对不同步长预测效果的影响.仿真结果表明,优化极限学习机预测误差可以用较小的网络规模获得很好的泛化能力.为流动不稳定性的实时预测提供了一种准确度较高的途径,其预测结果可以作为核动力系统操作员的参考.     

Residual Learning and Multi-Path Feature Fusion-Based Channel Estimation for Millimeter-Wave Massive MIMO System

Channel estimation is a challenging task in a millimeter-wave (mm Wave) massive multiple-input multiple-output (MIMO) system. The existing deep learning scheme, which learns the mapping from the input to the target channel, has great difficulty in estimating the exact channel state information (CSI). In this paper, we consider the quantized received measurements as a low-resolution image, and we adopt the deep learning-based image super-resolution technique to reconstruct the mm Wave channel. Specifically, we exploit a state-of-the-art channel estimation framework based on residual learning and multi-path feature fusion (RL-MFF-Net). Firstly, residual learning makes the channel estimator focus on learning high-frequency residual information between the quantized received measurements and the mm Wave channel, while abundant low-frequency information is bypassed through skip connections. Moreover, to address the estimator’s gradient dispersion problem, a dense connection is added to the residual blocks to ensure the maximum information flow between the layers. Furthermore, the underlying mm Wave channel local features extracted from different residual blocks are preserved by multi-path feature fusion. The simulation results demonstrate that the proposed scheme outperforms traditional methods as well as existing deep learning methods, especially in the low signal-to-noise-ration (SNR) region.     

Evolutionary games with two timescales

We consider a two timescale model of learning by economic agents wherein active or ‘ontogenetic’ learning by individuals takes place on a fast scale and passive or ‘phylogenetic’ learning by society as a whole on a slow scale, each affecting the evolution of the other. The former is modelled by the Monte Carlo dynamics of physics, while the latter is modelled by the replicator dynamics of evolutionary biology. Various quanlitative aspects of the dynamics are studied in some simple cases, both analytically and numerically, and its role as a useful modelling device is emphasized. rights reserved.     

Human cognitive paradigm and its application in semi-supervised learning

In many practical data mining applications such as web page classification, unlabeled training examples are readily available but labeled ones are fairly expensive to obtain. Therefore, semi-supervised learning algorithms such as Tri-training have attracted much attention. However, mislabeling the unlabeled data during the learning process is an inevitable problem and harms the performance improvement of the hypothesis. To solve this problem, a novel human cognitive paradigm is constructed for semi-supervised learning in this paper. In detail, based on local distribution of feature space, the majority voting scheme is substituted by an estimation of the probability of sample to belong to a certain class as an efficient strategy for data editing. It considers the form of the underlying probability distribution in the neighborhood of a point to identify and remove the mislabeled data. Validation of the proposed method is performed with extensive experiments. Results demonstrate that compared with Tri-training method, our method can more effectively and stably exploit unlabeled data to enhance the learning performance.     

基于深度学习的自适应光学波前传感技术研究综述

波前传感是自适应光学系统的重要组成部分,在地基大口径望远镜、激光大气传输、无线光通信、激光驱动核聚变等领域发挥了关键作用,同时也常应用于自由曲面的光学测量中。与此同时,深度学习作为一种较为通用的前沿技术,成功在计算机视觉、自然语言处理等众多领域取得了革命性进展。使用深度学习的方法改进自适应光学系统中的波前传感器,以期实现更精准的波前探测,以及适应更复杂的应用场景是自适应光学的发展趋势,也是深度学习应用领域的一个新课题。介绍了深度学习在自适应光学波前传感中的应用现状,主要分析了在相位反演波前传感器和哈特曼波前传感器中的研究特点,并在最后进行了总结和展望。     

Federated Quantum Machine Learning

Distributed training across several quantum computers could significantly improve the training time and if we could share the learned model, not the data, it could potentially improve the data privacy as the training would happen where the data is located. One of the potential schemes to achieve this property is the federated learning (FL), which consists of several clients or local nodes learning on their own data and a central node to aggregate the models collected from those local nodes. However, to the best of our knowledge, no work has been done in quantum machine learning (QML) in federation setting yet. In this work, we present the federated training on hybrid quantum-classical machine learning models although our framework could be generalized to pure quantum machine learning model. Specifically, we consider the quantum neural network (QNN) coupled with classical pre-trained convolutional model. Our distributed federated learning scheme demonstrated almost the same level of trained model accuracies and yet significantly faster distributed training. It demonstrates a promising future research direction for scaling and privacy aspects.     

Direct regressions for underwater acoustic source localization in fluctuating oceans

In this paper, we show the potential of machine learning regarding the task of underwater source localization through a fluctuating ocean. Underwater source localization is classically addressed under the angle of inversion techniques. However, because an inversion scheme is necessarily based on the knowledge of the environmental parameters, it may be not well adapted to a random and fluctuating underwater channel. Conversely, machine learning only requires using a training database, the environmental characteristics underlying the regression models. This makes machine learning adapted to fluctuating channels. In this paper, we propose to use non linear regressions for source localization in fluctuating oceans. The kernel regression as well as the local linear regression are compared to typical inversion techniques, namely Matched Field Beamforming and the algorithm MUSIC. Our experiments use both real tank-based and simulated data, introduced in the works of Real et al. Based on Monte Carlo iterations, we show that the machine learning approaches may outperform the inversion techniques.     

Task’s Choice: Pruning-Based Feature Sharing (PBFS) for Multi-Task Learning

In most of the existing multi-task learning (MTL) models, multiple tasks’ public information is learned by sharing parameters across hidden layers, such as hard sharing, soft sharing, and hierarchical sharing. One promising approach is to introduce model pruning into information learning, such as sparse sharing, which is regarded as being outstanding in knowledge transferring. However, the above method performs inefficiently in conflict tasks, with inadequate learning of tasks’ private information, or through suffering from negative transferring. In this paper, we propose a multi-task learning model (Pruning-Based Feature Sharing, PBFS) that merges a soft parameter sharing structure with model pruning and adds a prunable shared network among different task-specific subnets. In this way, each task can select parameters in a shared subnet, according to its requirements. Experiments are conducted on three benchmark public datasets and one synthetic dataset; the impact of the different subnets’ sparsity and tasks’ correlations to the model performance is analyzed. Results show that the proposed model’s information sharing strategy is helpful to transfer learning and superior to the several comparison models.     

高效选取线性独立的原子特征构建精确机器学习势函数

机器学习势由于具有与第一性原理计算相当的准确性,且低得多的计算成本,在原子模拟中极具前景. 然而原子机器学习势的可靠性、速度和可迁移性在很大程度上取决于原子构型的表示. 适当地选取用作机器学习程序输入的描述符是一个成功的机器学习表示的关键. 本文发展了一种简单有效的方法,可以基于训练数据固有的相关性,从大量待选的描述符中自动选取一组最佳的线性独立原子特征. 通过对几个具有较少冗余线性独立嵌入密度描述符的基准分子构建嵌入原子神经网络势的应用,证明了这种新方法的有效性和准确性. 该算法可以大大简化原子特征的初始选取,并极大地提高原子机器学习势的性能.     

疫情期间工科院校大学物理线上教学模式探究

为了适应新冠肺炎疫情时期线上教学要求,保证大学物理课程教学的顺利实施,本学期开始我校就对特殊时期的线上教学模式进行了探索.通过调研不同教学平台线上教学特点,最终决定采用钉钉和雨课堂作为主要的授课软件,学校网络教学平台作为教学资源上传平台,保证学生有系统的线上文件库.因为雨课堂的数据记录与分析功能,雨课堂被选为单元测试及课堂测试平台.通过在各平台实施“课前预习、课中直播讨论学习、课后复习”的全过程教学模式,不仅有效的开展了线上教学,还能采集到学生的学习数据,有效掌握学生的学习情况,为大学物理线上教学提供保障.     

Experimental investigation of the effects of the acoustical conditions in a simulated classroom on speech recognition and learning in children

The potential effects of acoustical environment on speech understanding are especially important as children enter school where students' ability to hear and understand complex verbal information is critical to learning. However, this ability is compromised because of widely varied and unfavorable classroom acoustics. The extent to which unfavorable classroom acoustics affect children's performance on longer learning tasks is largely unknown as most research has focused on testing children using words, syllables, or sentences as stimuli. In the current study, a simulated classroom environment was used to measure comprehension performance of two classroom learning activities: a discussion and lecture. Comprehension performance was measured for groups of elementary-aged students in one of four environments with varied reverberation times and background noise levels. The reverberation time was either 0.6 or 1.5 s, and the signal-to-noise level was either +10 or +7 dB. Performance is compared to adult subjects as well as to sentence-recognition in the same condition. Significant differences were seen in comprehension scores as a function of age and condition; both increasing background noise and reverberation degraded performance in comprehension tasks compared to minimal differences in measures of sentence-recognition.     

Co-Evolution of Predator-Prey Ecosystems by Reinforcement Learning Agents

The problem of finding adequate population models in ecology is important for understanding essential aspects of their dynamic nature. Since analyzing and accurately predicting the intelligent adaptation of multiple species is difficult due to their complex interactions, the study of population dynamics still remains a challenging task in computational biology. In this paper, we use a modern deep reinforcement learning (RL) approach to explore a new avenue for understanding predator-prey ecosystems. Recently, reinforcement learning methods have achieved impressive results in areas, such as games and robotics. RL agents generally focus on building strategies for taking actions in an environment in order to maximize their expected returns. Here we frame the co-evolution of predators and preys in an ecosystem as allowing agents to learn and evolve toward better ones in a manner appropriate for multi-agent reinforcement learning. Recent significant advancements in reinforcement learning allow for new perspectives on these types of ecological issues. Our simulation results show that throughout the scenarios with RL agents, predators can achieve a reasonable level of sustainability, along with their preys.     

Conditional Deep Gaussian Processes: Empirical Bayes Hyperdata Learning

It is desirable to combine the expressive power of deep learning with Gaussian Process (GP) in one expressive Bayesian learning model. Deep kernel learning showed success as a deep network used for feature extraction. Then, a GP was used as the function model. Recently, it was suggested that, albeit training with marginal likelihood, the deterministic nature of a feature extractor might lead to overfitting, and replacement with a Bayesian network seemed to cure it. Here, we propose the conditional deep Gaussian process (DGP) in which the intermediate GPs in hierarchical composition are supported by the hyperdata and the exposed GP remains zero mean. Motivated by the inducing points in sparse GP, the hyperdata also play the role of function supports, but are hyperparameters rather than random variables. It follows our previous moment matching approach to approximate the marginal prior for conditional DGP with a GP carrying an effective kernel. Thus, as in empirical Bayes, the hyperdata are learned by optimizing the approximate marginal likelihood which implicitly depends on the hyperdata via the kernel. We show the equivalence with the deep kernel learning in the limit of dense hyperdata in latent space. However, the conditional DGP and the corresponding approximate inference enjoy the benefit of being more Bayesian than deep kernel learning. Preliminary extrapolation results demonstrate expressive power from the depth of hierarchy by exploiting the exact covariance and hyperdata learning, in comparison with GP kernel composition, DGP variational inference and deep kernel learning. We also address the non-Gaussian aspect of our model as well as way of upgrading to a full Bayes inference.     

基于内部学习动机的中学物理课堂引入

分析了认知学习理论中的学习动机的构成,认为在中学物理课堂引入环节教学,通过创设问题情境,激发学习需要;随之剖析问题情境,帮助学生形成学习期待,可有效地激励出其物理学习动机;然后及时运用所学知识消除问题情境,以求达到进一步强化学习动机的目的.