基于One-Shot聚合自编码器的图表示学习

自编码器（AE）是一种高效的图数据表示学习模型,但大多数图自编码器（GAE）为浅层模型,其效率会随着隐藏层的增加而降低。针对上述问题,提出基于One-Shot聚合（OSA）和指数线性（ELU）函数的GAE模型OSA-GAE和图变分自编码器模型OSA-VGAE。首先,利用多层图卷积网络（GCN）构建编码器,并引入OSA和ELU函数;然后,在解码阶段使用内积解码器恢复图的拓扑结构;此外,为了防止模型训练过程中的参数过拟合,在损失函数中引入正则化项。实验结果表明,OSA和ELU函数可以有效提高深层GAE的性能,改善模型的梯度信息传递。在使用6层GCN时,基准引文数据集PubMed的链接预测任务中,深层OSA-VGAE相较于原始的VGAE在ROC曲线下的面积（AUC）和平均精度（AP）上分别提升了8.67和6.85个百分点,深层OSA-GAE相较于原始的GAE在AP和AUC上分别提升了6.82和4.39个百分点。     

Real-time one-shot learning gesture recognition based on lightweight 3D Inception-ResNet with separable convolutions

Pattern Analysis and Applications - Gesture recognition is a popular research field in computer vision and the application of deep neural networks greatly improves its performance. However, the...     

Improved one-shot learning for feedforward associative memorieswith application to composite pattern association

The local identical index (LII) associative memory (AM) proposed by the authors in a previous paper is a one-shot feedforward structure designed to exhibit no spurious attractors. In this paper we relax the latter design constraint in exchange for enlarged basins of attraction and we develop a family of modified LII AM networks that exhibit improved performance, particularly in memorizing highly correlated patterns. The new algorithm meets the requirement of no spurious attractors only in a local sense. Finally, we show that the modified LII family of networks can accommodate composite patterns of any size by storing (memorizing) only the basic (prime) prototype patterns. The latter property translates to low learning complexity and a simple network structure with significant memory savings. Simulation studies and comparisons illustrate and support the the optical developments.     

A morphable template framework for robot learning by demonstration: Integrating one-shot and incremental learning approaches

Robot learning by demonstration is key to bringing robots into daily social environments to interact with and learn from human and other agents. However, teaching a robot to acquire new knowledge is a tedious and repetitive process and often restrictive to a specific setup of the environment. We propose a template-based learning framework for robot learning by demonstration to address both generalisation and adaptability. This novel framework is based upon a one-shot learning model integrated with spectral clustering and an online learning model to learn and adapt actions in similar scenarios. A set of statistical experiments is used to benchmark the framework components and shows that this approach requires no extensive training for generalisation and can adapt to environmental changes flexibly. Two real-world applications of an iCub humanoid robot playing the tic-tac-toe game and soldering a circuit board are used to demonstrate the relative merits of the framework.     

Maximal width learning of binary functions

This paper concerns learning binary-valued functions defined on R, and investigates how a particular type of ‘regularity’ of hypotheses can be used to obtain better generalization error bounds. We derive error bounds that depend on the sample width (a notion analogous to that of sample margin for real-valued functions). This motivates learning algorithms that seek to maximize sample width.     

Real-time SLAM relocalization with online learning of binary feature indexing

A visual simultaneous localization and mapping (SLAM) system usually contains a relocalization module to recover the camera pose after tracking failure. The core of this module is to establish correspondences between map points and key points in the image, which is typically achieved by local image feature matching. Since recently emerged binary features have orders of magnitudes higher extraction speed than traditional features such as scale invariant feature transform, they can be applied to develop a real-time relocalization module once an efficient method of binary feature matching is provided. In this paper, we propose such a method by indexing binary features with hashing. Being different from the popular locality sensitive hashing, the proposed method constructs the hash keys by an online learning process instead of pure randomness. Specifically, the hash keys are trained with the aim of attaining uniform hash buckets and high collision rates of matched feature pairs, which makes the method more efficient on approximate nearest neighbor search. By distributing the online learning into the simultaneous localization and mapping process, we successfully apply the method to SLAM relocalization. Experiments show that camera poses can be recovered in real time even when there are tens of thousands of landmarks in the map.     

Optimizing conditions for computer-assisted anatomical learning

An experiment evaluated the impact of two typical features of virtual learning environments on anatomical learning for users of differing visuo-spatial ability. The two features studied are computer-implemented stereopsis (the spatial information that is based on differences in visual patterns projected in both eyes) and interactivity (the possibility to actively and continuously change one's view of computer-mediated objects). Participants of differing visuo-spatial ability learned about human abdominal organs via anatomical three-dimensional (3D) reconstructions using either a stereoptic study phase (involving stereopsis and interactivity) or using a biocular study phase that involved neither stereopsis nor interactivity. Subsequent tests assessed the acquired knowledge in tasks involving (a) identification of anatomical structures in anatomical 2D cross-sections (i.e. typical Computed Tomography pictures) in an identification task, and (b) localization of these cross-sections in a frontal view of the anatomy in a localization task. The results show that the stereoptic group performed significantly better on both tasks and that participants of low visuo-spatial ability benefited more from the stereoptic study phase than those of high visuo-spatial ability.     

Optimizing the binary discriminant function in change detection applications

Binary discriminant functions are often used to identify changed area through time in remote sensing change detection studies. Traditionally, a single change-enhanced image has been used to optimize the binary discriminant function with a few (e.g., 5-10) discrete thresholds using a trial-and-error method. Im et al. [Im, J., Rhee, J., Jensen, J. R., & Hodgson, M. E. (2007). An automated binary change detection model using a calibration approach. Remote Sensing of Environment, 106, 89-105] developed an automated calibration model for optimizing the binary discriminant function by autonomously testing thousands of thresholds. However, the automated model may be time-consuming especially when multiple change-enhanced images are used as inputs together since the model is based on an exhaustive search technique. This paper describes the development of a computationally efficient search technique for identifying optimum threshold(s) in a remote sensing spectral search space. The new algorithm is based on “systematic searching.” Two additional heuristic optimization algorithms (i.e., hill climbing, simulated annealing) were examined for comparison. A case study using QuickBird and IKONOS satellite imagery was performed to evaluate the effectiveness of the proposed algorithm. The proposed systematic search technique reduced the processing time required to identify the optimum binary discriminate function without decreasing accuracy. The other two optimizing search algorithms also reduced the processing time but failed to detect a global maxima for some spectral features.     

Optimizing task scheduling in human-robot collaboration with deep multi-agent reinforcement learning

Human-Robot Collaboration (HRC) presents an opportunity to improve the efficiency of manufacturing processes. However, the existing task planning approaches for HRC are still limited in many ways, e.g., co-robot encoding must rely on experts’ knowledge and the real-time task scheduling is applicable within small state-action spaces or simplified problem settings. In this paper, the HRC assembly working process is formatted into a novel chessboard setting, in which the selection of chess piece move is used to analogize to the decision making by both humans and robots in the HRC assembly working process. To optimize the completion time, a Markov game model is considered, which takes the task structure and the agent status as the state input and the overall completion time as the reward. Without experts’ knowledge, this game model is capable of seeking for correlated equilibrium policy among agents with convergency in making real-time decisions facing a dynamic environment. To improve the efficiency in finding an optimal policy of the task scheduling, a deep-Q-network (DQN) based multi-agent reinforcement learning (MARL) method is applied and compared with the Nash-Q learning, dynamic programming and the DQN-based single-agent reinforcement learning method. A height-adjustable desk assembly is used as a case study to demonstrate the effectiveness of the proposed algorithm with different number of tasks and agents.     

A one-shot bargaining strategy for dealing with multifarious opponents

Bargaining is an effective paradigm to solve the problem of resource allocation. The consideration of factors such as bounded rationality of negotiators, time constraints, incomplete information, and complexity of dynamic environment make the design of optimal strategy for one-shot bargaining much tougher than the situation that all bargainers are assumed to be absolutely rational. Lots of prediction-based strategies have been explored either based on assuming a finite number of models for opponents, or focusing on the prediction of opponent’s reserve price, deadline, or the probabilities of different behaviors. Following the methods of estimating opponent’s private information, this paper gives a strategy which improves the BLGAN strategy to adapt to various possible bargaining situations and deal with multifarious opponents. In addition, this paper compares the improved BLGAN strategy with related work. Experimental results show that the improved BLGAN strategy can outperform related ones when faced with various opponents, especially the agents who frequently change their strategies for anti-learning.     

Online shape learning using binary search trees

In this paper we propose an online shape learning algorithm based on the self-balancing binary search tree data structure for the storage and retrieval of shape templates. This structure can also be used for classification purposes. We introduce a similarity measure with which we can make decisions on how to traverse the tree and even backtrack through the search path to find more candidate matches. Then we describe every basic operation a binary search tree can perform adapted to such a tree of shapes. Note that as a property of binary search trees, all operations can be performed in O(logn)$O(\log n)$ time and are very efficient. Finally, we present experimental data evaluating the performance of the proposed algorithm and demonstrating the suitability of this data structure for the purpose it was designed to serve.     

The geometrical learning of binary neural networks

In this paper, the learning algorithm called expand-and-truncate learning (ETL) is proposed to train multilayer binary neural networks (BNN) with guaranteed convergence for any binary-to-binary mapping. The most significant contribution of this paper is the development of a learning algorithm for three-layer BNN which guarantees the convergence, automatically determining a required number of neurons in the hidden layer. Furthermore, the learning speed of the proposed ETL algorithm is much faster than that of backpropagation learning algorithm in a binary field. Neurons in the proposed BNN employ a hard-limiter activation function, with only integer weights and integer thresholds. Therefore, this will greatly facilitate actual hardware implementation of the proposed BNN using currently available digital VLSI technology     

Semi-supervised multi-view binary learning for large-scale image clustering

Applied Intelligence - Large-scale image clustering has attracted sustained attention in machine learning. The traditional methods based on real value representation often suffer from the data...     

Deep learning compact binary codes for fingerprint indexing

With the rapid growth in fingerprint databases, it has become necessary to develop excellent fingerprint indexing to achieve efficiency and accuracy. Fingerprint indexing has been widely studied with real-valued features, but few studies focus on binary feature representation, which is more suitable to identify fingerprints efficiently in large-scale fingerprint databases. In this study, we propose a deep compact binary minutia cylinder code (DCBMCC) as an effective and discriminative feature representation for fingerprint indexing. Specifically, the minutia cylinder code (MCC), as the state-of-the-art fingerprint representation, is analyzed and its shortcomings are revealed. Accordingly, we propose a novel fingerprint indexing method based on deep neural networks to learn DCBMCC. Our novel network restricts the penultimate layer to directly output binary codes. Moreover, we incorporate independence, balance, quantization-loss-minimum, and similarity-preservation properties in this learning process. Eventually, a multi-index hashing (MIH) based fingerprint indexing scheme further speeds up the exact search in the Hamming space by building multiple hash tables on binary code substrings. Furthermore, numerous experiments on public databases show that the proposed approach is an outstanding fingerprint indexing method since it has an extremely small error rate with a very low penetration rate.     

Hybrid learning of Bayesian multinets for binary classification

We propose a scoring criterion, named mixture-based factorized conditional log-likelihood (mfCLL), which allows for efficient hybrid learning of mixtures of Bayesian networks in binary classification tasks. The learning procedure is decoupled in foreground and background learning, being the foreground the single concept of interest that we want to distinguish from a highly complex background. The overall procedure is hybrid as the foreground is discriminatively learned, whereas the background is generatively learned. The learning algorithm is shown to run in polynomial time for network structures such as trees and consistent κ-graphs. To gauge the performance of the mfCLL scoring criterion, we carry out a comparison with state-of-the-art classifiers. Results obtained with a large suite of benchmark datasets show that mfCLL-trained classifiers are a competitive alternative and should be taken into consideration.     

Optimizing area under the ROC curve using semi-supervised learning

Receiver operating characteristic (ROC) analysis is a standard methodology to evaluate the performance of a binary classification system. The area under the ROC curve (AUC) is a performance metric that summarizes how well a classifier separates two classes. Traditional AUC optimization techniques are supervised learning methods that utilize only labeled data (i.e., the true class is known for all data) to train the classifiers. In this work, inspired by semi-supervised and transductive learning, we propose two new AUC optimization algorithms hereby referred to as semi-supervised learning receiver operating characteristic (SSLROC) algorithms, which utilize unlabeled test samples in classifier training to maximize AUC. Unlabeled samples are incorporated into the AUC optimization process, and their ranking relationships to labeled positive and negative training samples are considered as optimization constraints. The introduced test samples will cause the learned decision boundary in a multi-dimensional feature space to adapt not only to the distribution of labeled training data, but also to the distribution of unlabeled test data. We formulate the semi-supervised AUC optimization problem as a semi-definite programming problem based on the margin maximization theory. The proposed methods SSLROC1 (1-norm) and SSLROC2 (2-norm) were evaluated using 34 (determined by power analysis) randomly selected datasets from the University of California, Irvine machine learning repository. Wilcoxon signed rank tests showed that the proposed methods achieved significant improvement compared with state-of-the-art methods. The proposed methods were also applied to a CT colonography dataset for colonic polyp classification and showed promising results.¹     

Associative memory with dynamic synapses

We have examined a role of dynamic synapses in the stochastic Hopfield-like network behavior. Our results demonstrate an appearance of a novel phase characterized by quick transitions from one memory state to another. The network is able to retrieve memorized patterns corresponding to classical ferromagnetic states but switches between memorized patterns with an intermittent type of behavior. This phenomenon might reflect the flexibility of real neural systems and their readiness to receive and respond to novel and changing external stimuli.     

Coincidence detection with dynamic synapses

Recent experimental findings show that the efficacy of transmission in cortical synapses depends on presynaptic activity. In most neural models, however, the synapses are regarded as static entities where this dependence is not included. We study the role of activity-dependent (dynamic) synapses in neuronal responses to temporal patterns of afferent activity. Our results demonstrate that, for suitably chosen threshold values, dynamic synapses are capable of coincidence detection (CD) over a much larger range of frequencies than static synapses. The phenomenon appears to be valid for an integrate-and-fire as well as a Hodgkin-Huxley neuron and various types of CD tasks.     

Principal motion components for one-shot gesture recognition

This paper introduces principal motion components (PMC), a new method for one-shot gesture recognition. In the considered scenario a single training video is available for each gesture to be recognized, which limits the application of traditional techniques (e.g., HMMs). In PMC, a 2D map of motion energy is obtained per each pair of consecutive frames in a video. Motion maps associated to a video are processed to obtain a PCA model, which is used for recognition under a reconstruction-error approach. The main benefits of the proposed approach are its simplicity, easiness of implementation, competitive performance and efficiency. We report experimental results in one-shot gesture recognition using the ChaLearn Gesture Dataset; a benchmark comprising more than 50,000 gestures, recorded as both RGB and depth video with a Kinect?camera. Results obtained with PMC are competitive with alternative methods proposed for the same data set.     

Image preprocessing with dynamic synapses

Different algorithms suitable for a specific class of picture were developed for image processing. We will represent the filtering capability of a spiking neural network based on dynamic synapses. For this intention we chose an x-ray image of the human coronary trees and another noisy image. In other words the task at hand is to show how accurately such a network is able to store various aspects (object/background) of stimulus in the variables which describe dynamic of synaptic response. The behavior of these synapses influences the effective connection in the network in a short time-scale. Such a network has a low activity and a balanced behavior. Dynamic synapses are able to adjust their behavior by fast changing stimuli. These synapses retain the information in the variables, such as potential and time.