Dynamic proximity of spatio-temporal sequences

Recurrent networks can generate spatio-temporal neural sequences of very large cycles, having an apparent random behavior. Nonetheless a proximity measure between these sequences may be defined through comparison of the synaptic weight matrices that generate them. Following the dynamic neural filter (DNF) formalism we demonstrate this concept by comparing teacher and student recurrent networks of binary neurons. We show that large sequences, providing a training set well exceeding the Cover limit, allow for good determination of the synaptic matrices. Alternatively, assuming the matrices to be known, very fast determination of the biases can be achieved. Thus, a spatio-temporal sequence may be regarded as spatio-temporal encoding of the bias vector. We introduce a linear support vector machine (SVM) variant of the DNF in order to specify an optimal weight matrix. This approach allows us to deal with noise. Spatio-temporal sequences generated by different DNFs with the same number of neurons may be compared by calculating correlations of the synaptic matrices of the reconstructed DNFs. Other types of spatio-temporal sequences need the introduction of hidden neurons, and/or the use of a kernel variant of the SVM approach. The latter is being defined as a recurrent support vector network (RSVN).     

Generating Fluttering Patterns with Low Autocorrelation for Coded Exposure Imaging

The performance of coded exposure imaging critically depends on finding good binary sequences. Previous coded exposure imaging methods have mostly relied on random searching to find the binary codes, but that approach can easily fail to find good long sequences, due to the exponentially expanding search space. In this paper, we present two algorithms for generating the binary sequences, which are especially well suited for generating short and long binary sequences, respectively. We show that the concept of low autocorrelation binary sequences, which has been successfully exploited in the field of information theory, can be applied to generate shutter fluttering patterns. We also propose a new measure for good binary sequences. Based on the new measure, we introduce two new algorithms for coded exposure imaging - a modified Legendre sequence method and a memetic algorithm. Experiments using both synthetic and real data show that our new algorithms consistently generate better binary sequences for the coded exposure problem, yielding better deblurring and resolution enhancement results compared to previous methods of generating the binary codes.     

基于神经网络与图象处理技术的油水层综合判别

提出了一种基于神经网络与图象处理技术相结合的油水层综合判别方法。首先将数字化测井曲线和地层参数经预处理转化为二值点阵图象模式,经过点阵数据编码压缩提取和记忆曲线所表征的地层模式特征,然后利用BP算法与遗传算法相结合的方法训练多层前馈神经网络。所得神经网络稳定、学习收敛速度快,同时有很强的记忆能力和推广能力,此模型一水层综合判别问题具有良好的适应性。通过对大庆油田采油八厂升平油田葡萄花油层8口井的资料处理,取得了很好的效果。     

字典学习联合粒子滤波鲁棒跟踪

针对运动目标鲁棒跟踪问题,提出一种基于离线字典学习的视频目标跟踪鲁棒算法。采用字典编码方式提取目标的局部区域描述符,随后通过训练分类器将跟踪问题转化为背景和前景分类问题,最终通过粒子滤波对物体位置进行估计实现跟踪。该算法能够有效解决由于光照变化、背景复杂、快速运动、遮挡产生的跟踪困难。经过不同图像序列的实验对比表明,与现有方法相比,本文算法的鲁棒性较高。     

Self-organizing dual coding based on spike-time-dependent plasticity

It has been a matter of debate how firing rates or spatiotemporal spike patterns carry information in the brain. Recent experimental and theoretical work in part showed that these codes, especially a population rate code and a synchronous code, can be dually used in a single architecture. However, we are not yet able to relate the role of firing rates and synchrony to the spatiotemporal structure of inputs and the architecture of neural networks. In this article, we examine how feedforward neural networks encode multiple input sources in the firing patterns. We apply spike-time-dependent plasticity as a fundamental mechanism to yield synaptic competition and the associated input filtering. We use the Fokker-Planck formalism to analyze the mechanism for synaptic competition in the case of multiple inputs, which underlies the formation of functional clusters in downstream layers in a self-organizing manner. Depending on the types of feedback coupling and shared connectivity, clusters are independently engaged in population rate coding or synchronous coding, or they interact to serve as input filters. Classes of dual codings and functional roles of spike-time-dependent plasticity are also discussed.     

On the Learnability of Disjunctive Normal Form Formulas

We present two related results about the learnability of disjunctive normal form (DNF) formulas. First we show that a common approach for learning arbitrary DNF formulas requires exponential time. We then contrast this with a polynomial time algorithm for learning most (rather than all) DNF formulas. A natural approach for learning boolean functions involves greedily collecting the prime implicants of the hidden function. In a seminal paper of learning theory, Valiant demonstrated the efficacy of this approach for learning monotone DNF, and suggested this approach for learning DNF. Here we show that no algorithm using such an approach can learn DNF in polynomial time. We show this by constructing a counterexample DNF formula which would force such an algorithm to take exponential time. This counterexample seems to capture much of what makes DNF hard to learn, and thus is useful to consider when evaluating the run-time of a proposed DNF learning algorithm. This hardness result, as well as other hardness results for learning DNF, relies on the construction of particular hard-to-learn formulas, formulas that appear to be relatively rare. This raises the question of whether most DNF formulas are learnable. For certain natural definitions of most DNF formulas, we answer this question affirmatively.     

Ternary Bradley-Terry model-based decoding for?multi-class classification and its extensions

A multi-class classifier based on the Bradley-Terry model predicts the multi-class label of an input by combining the outputs from multiple binary classifiers, where the combination should be a priori designed as a code word matrix. The code word matrix was originally designed to consist of +1 and ?1 codes, and was later extended into deal with ternary code {+1,0,?1}, that is, allowing 0 codes. This extension has seemed to work effectively but, in fact, contains a problem: a binary classifier forcibly categorizes examples with 0 codes into either +1 or ?1, but this forcible decision makes the prediction of the multi-class label obscure. In this article, we propose a Boosting algorithm that deals with three categories by allowing a ??don??t care?? category corresponding to 0 codes, and present a modified decoding method called a ??ternary?? Bradley-Terry model. In addition, we propose a couple of fast decoding schemes that reduce the heavy computation by the existing Bradley-Terry model-based decoding.     

Synaptic dynamics in analog VLSI

Synapses are crucial elements for computation and information transfer in both real and artificial neural systems. Recent experimental findings and theoretical models of pulse-based neural networks suggest that synaptic dynamics can play a crucial role for learning neural codes and encoding spatiotemporal spike patterns. Within the context of hardware implementations of pulse-based neural networks, several analog VLSI circuits modeling synaptic functionality have been proposed. We present an overview of previously proposed circuits and describe a novel analog VLSI synaptic circuit suitable for integration in large VLSI spike-based neural systems. The circuit proposed is based on a computational model that fits the real postsynaptic currents with exponentials. We present experimental data showing how the circuit exhibits realistic dynamics and show how it can be connected to additional modules for implementing a wide range of synaptic properties.     

运动序列的时空结构特征表示模型

运动序列是一种与运动信号相关的多维时间序列,各个维度序列之间具有高耦合性的特点。现有的多维序列表征方法大多基于维度间相互独立的假设或缺乏可解释性,为此,提出一种适用于运动序列的时空结构特征表示模型及其两阶段构造方法。首先,基于空间变化事件的转换方法,将多维时间序列变换成一维事件序列,以保存序列中的空间结构特性。接着,定义了一种时空结构特征的无监督挖掘算法。基于新定义的表示度度量,该算法从事件序列中提取一组具有代表性的低冗余变长事件元组为时空结构特征。在多个人类行为识别数据集上的实验结果表明,与现有多维时间序列表示方法相比,新模型的特征集更具代表性,在运动序列模式识别领域可以有效提升分类精度。     

基于耦合映像格子模型的时空混沌二值序列及其性能分析

基于耦合映像格子模型。给出了一种时空混沌二值序列的产生方法,并对其性能进行了详细分析。结果表明,该种时空混沌二值序列具有十分理想的随机性和相关性。此外,混沌序列容易产生和控制,具有线性复杂度高、对参数敏感等特性,因此特别适合于在保密通信和密码学等诸多领域中应用。     

DNA-based algorithms for learning Boolean formulae

We apply a DNA-based massively parallel exhaustive search to solving the computational learning problems of DNF (disjunctive normal form) Boolean formulae. Learning DNF formulae from examples is one of the most important open problems in computational learning theory and the problem of learning 3-term DNF formulae is known as intractable if RP NP. We propose new methods to encode any k-term DNF formula to a DNA strand, evaluate the encoded DNF formula for a truth-value assignment by using hybridization and primer extension with DNA polymerase, and find a consistent DNF formula with the given examples. By employing these methods, we show that the class of k-term DNF formulae (for any constant k) and the class of general DNF formulae are efficiently learnable on DNA computer.Second, in order for the DNA-based learning algorithm to be robust for errors in the data, we implement the weighted majority algorithm on DNA computers, called DNA-based majority algorithm via amplification (DNAMA), which take a strategy of ``amplifying' the consistent (correct) DNA strands. We show a theoretical analysis for the mistake bound of the DNA-based majority algorithm via amplification, and imply that the amplification to ``double the volumes' of the correct DNA strands in the test tube works well.     

基于字典学习的粒子滤波鲁棒跟踪算法

字典学习广泛应用于图像去噪、图像分类等领域,但是将离线字典训练如何应用于视频目标跟踪的研究较少。本文采用一种字典编码方法提取目标的局部区域描述符,通过训练分类器将跟踪问题转化为背景和前景二值分类问题,并通过粒子滤波对物体位置进行估计实现跟踪。不同图像序列的实验结果表明,与现有的方法相比本文的算法具有较好的鲁棒性。     

Increase of the speed of operation of scalar neural network tree when solving the nearest neighbor search problem in binary space of large dimension

In the binary space of large dimension we analyze the nearest neighbor search problem where the required point is a distorted version of one of the patterns. Previously it was shown that the only algorithms able to solve the set problem are the exhaustive search and the neural network search tree. For the given problem the speed of operation of the last algorithm is dozens of times larger comparing with the exhaustive search. Moreover, in the case of large dimensions the neural network tree can be regarded as an accurate algorithm since the probability of its error is so small that cannot be measured. In the present publication, we propose a modification of the scalar neural network tree allowing the speeding of the algorithm’s operation up to hundred times without losses in its reliability.     

A new chain-coding algorithm for binary images using run-length codes

This paper presents a new algorithm for converting a binary image into chain codes using its run-length codes. The basic idea of conventional chain-coding algorithm is to follow boundary pixels by convolving a 3 × 3 window with the image and to sequentially generate chain codes. The proposed algorithm has two phases, namely run-length coding and chain-code generation. We use connectivity information between runs as well as their coordinates in the phase of run-length coding. In the second phase (chain-code generation) the connectivity information extracted in the first phase is utilized for sequentially tracking runs containing the boundary pixels to be followed. This algorithm has an advantage that we can detect easily the inclusion relationship between boundaries at the same time as chain-code generation.     

A special class of quasi-cyclic low-density parity-check codes based on repetition codes and permutation matrices

We propose a new ensemble of binary low-density parity-check codes with paritycheck matrices based on repetition codes and permutation matrices. The proposed class of codes is a subensemble of quasi-cyclic codes. For the constructed ensemble, we obtain minimum distance estimates. We present simulation results for the proposed code constructions under the (Sum-Product) iterative decoding algorithm for transmission over an additive white Gaussian noise channel using binary phase-shift keying.     

Minority codes with improved embedding efficiency for large payloads

Improving embedding efficiency is important for watermarking with large payloads. In this paper, we propose a coding scheme named Minority codes by observing the relationship between the populations and positions of the binary bits in a sequence whose length is an odd number. In such a sequence, there is always a bit value, whose population is minority. We observe that the positions of the minority bit in two complement sequences are the same. By using this property, we create Minority codes whose codebook is composed of a pair of complement sequences for each entry. Minority codes can be combined with watermarking algorithms to improve embedding efficiency, because we can always identify the codeword that causes fewer embedding changes according to the host image and the watermarking method. The performance of Minority codes is analyzed theoretically and supported experimentally. The complexity of Minority codes is quite low and suitable for large payloads.     

Evolving Output Codes for Multiclass Problems

In this paper, we propose an evolutionary approach to the design of output codes for multiclass pattern recognition problems. This approach has the advantage of taking into account the different aspects that are relevant for a code matrix to achieve a good performance. We define a fitness function made up of five terms that refer to overall classifier accuracy, binary classifiers' accuracy, classifiers' diversity, minimum Hamming distance among codewords, and margin of classification. These five factors have not been considered together in previous works. We perform a study of these five terms to obtain a fitness function with three of them. We test our approach on 27 datasets from the UCI Machine Learning Repository, using three different base learners: C4.5, neural networks, and support vector machines. We show a better performance than most of the current standard methods, namely, randomly generated codes with approximately equal random split, codes designed using a CHC algorithm, and one-vs-all and one-vs-one methods.     

Boolean functions with long prime implicants

In this short note we introduce a hierarchy of classes of Boolean functions, where each class is defined by the minimum allowed length of prime implicants of the functions in the class. We show that for a given DNF and a given class in the hierarchy, it is possible to test in polynomial time whether the DNF represents a function from the given class. For the first class in the hierarchy we moreover present a polynomial time algorithm which for a given input DNF outputs a shortest logically equivalent DNF, i.e. a shortest DNF representation of the underlying function. This class is therefore a new member of a relatively small family of classes for which the Boolean minimization problem can be solved in polynomial time. For the second class and higher classes in the hierarchy we show that the Boolean minimization problem can be approximated within a constant factor.     

Efficient and Robust Path Openings Using the Scale-Invariant Rank Operator

Some basic properties of a slightly generalized version of the scale-invariant rank operator are given, and it is shown how this operator can be used to create a nearly scale-invariant generalization of path openings that is robust to noise. Efficient algorithms are given for sequences and directed acyclic graphs with binary values, as well as sequences with real (greyscale) values. An algorithm is also given for directed acyclic graphs with real weights. It is shown that the given algorithms might be extended even further by allowing for scores based on a totally ordered semigroup.