Pattern Retrieval Research Articles

Signatures of Associative Memory Behavior in a Multimode Dicke Model.

Dicke-like models can describe a variety of physical systems, such as atoms in a cavity or vibrating ion chains. In equilibrium these systems often feature a radical change in their behavior when switching from weak to strong spin-boson interaction. This usually manifests in a transition from a "dark" to a "superradiant" phase. However, understanding the out-of-equilibrium physics of these models is extremely challenging, and even more so for strong spin-boson coupling. Here we show that the nonequilibrium strongly interacting multimode Dicke model can mimic some fundamental properties of an associative memory-a system which permits the recognition of patterns, such as letters of an alphabet. Patterns are encoded in the couplings between spins and bosons, and we discuss the dynamics of the spins from the perspective of pattern retrieval in associative memory models. We identify two phases, a "paramagnetic" and a "ferromagnetic" one, and a crossover behavior between these regimes. The "ferromagnetic" phase is reminiscent of pattern retrieval. We highlight similarities and differences with the thermal dynamics of a Hopfield associative memory and show that indeed elements of "machine learning behavior" emerge in the strongly coupled multimode Dicke model.

Pattern retrieval of yarn‐dyed plaid fabric based on modified interactive genetic algorithm

AbstractWith the development of perceptual consumption, consumers sometimes cannot explicitly describe the purchase demands or only based on impression, like the color perception and aesthetic experience. Based on the consumer's expression, it is difficult to design a new fabric by repeated proofing to meet the consumer's demands. To retrieve the existing patterns incorporating human intuition and emotion, this study proposed a novel pattern retrieval method of yarn‐dyed plaid fabric using modified interactive genetic algorithm. Each pattern was encoded based on the design elements and visual features were extracted to bridge the semantic gap between the designer and the consumer. Survival of the fittest and two special mutation operators, addition and deletion, were designed to increase the diversity of the generations. During the evolution, the generated patterns were replaced by the most similar patterns in the database based on visual features. Experimental results showed that the proposed scheme is feasible and effective to extract the consumer's preferences and retrieve satisfactory patterns, helping the factory obtain the process sheet to guide production and save labor and material resources.

Neuromorphic Computing in Autoassociative Memory with a Regular Spiking Neuron Model

Digital Spiking Silicon Neuron (DSSN) model is a qualitative neuron model specifically designed for efficient digital circuit implementation which exhibits high biological plausibility. In this study we analyzed the behavior of an autoassociative memory composed of 3-variable DSSN model which has a slow negative feedback variable that models the effect of slow ionic currents responsible for Spike Frequency Adaptation (SFA). We observed the network dynamics by altering the strength of SFA which is known to be dependent on Acetylcholine volume, together with the magnitude of neuronal interaction. By altering these parameters, we obtained various pattern retrieval dynamics, such as chaotic transitions within stored patterns or stable and high retrieval performance. In the end, we discuss potential applications of the obtained results for neuromorphic computing.

Efficient Hybrid CMOS/Memristor Implementation of Bidirectional Associative Memory Using Passive Weight Array

The primacy of memristors for the realization of the synaptic circuit makes this element a promising component for implementing Artificial Neural Networks (ANN). However, integrating memristors with other ANN components is still an open area of research. Therefore, a circuit-level hybrid CMOS/Memristor architecture for implementing Bidirectional Associative Memory (BAM) is proposed in this paper. A passive bi-directional pre-differentiation crossbar architecture has been utilized and applied for the realization of synaptic weights. A compact, low voltage CMOS neuron that consumes a low amount of power is also designed. The noise tolerance of the proposed hardware is evaluated for typical networks with different sizes and a number of stored patterns. It is demonstrated that the hardware pattern retrieval rate is similar to its software-based counterpart. The influence of memristor non-idealities on BAM is also investigated for the first time, and a practical method has been introduced to deal with this issue. In comparison to the available circuit-level hardware implementation of memristive associative memories, the proposed circuit consumes significantly less power and retrieves the stored patterns notably faster.

Repeat pattern segmentation of print fabric based on adaptive template matching

Patterned fabrics are generally constructed from the periodic repetition of a primitive pattern unit. Repeat pattern segmentation of printed fabrics has a very significant impact on the pattern retrieval and pattern defect detection. In this paper, we propose a new approach for repeat pattern segmentation by employing the adaptive template matching method. In contrast to the traditional method for template matching, the proposed algorithm first selects an adaptive size template image in the repeat pattern image based on the size of the original image and its local maximum edge density. Then it uses the sum of absolute differences as the matching features to identify the matched regions in the original image, and the minimum envelope border of the primitive pattern, typically as a parallelogram, can be determined from the results of the four adjacent matched templates. Finally, image traversal base on the obtained parallelogram is implemented over the original image using minimum information loss theory to produce a well-segmented primitive pattern with a complete edge structure. The results from the experiments conducted using an extensive database of real fabric images show that the proposed algorithm has the advantage of rotation invariance and scaling invariance and will not be affected when the background or foreground color is changed.

Mining patterns in open source software using software metrics and neural network models

As the complexity of a system increases, the need to develop a prototype for the problems becomes more and more prominent. To take care of different plan issues, it is seen that structure design finds a superior answer for a large number of repetitive plan issues. For the most part, design-level documents are indicated utilising semi-formal documentation, for example, unified modelling language (UML) diagrams. In any case, these kinds of semi-formal documentation lead to ambiguities and irregularities. In this paper, design level documents are retrieved by using the idea of program metrics. The proposed methodology retrieves the reusable documents from the open-source software such as PMD and JUnit. The presented method uses three kinds of neural network models to analyse the effectiveness of the pattern retrieval approach.

Mining Patterns in Open Source Software using Software Metrics and Neural Network Models

As the complexity of a system increases, the need to develop a prototype for the problems becomes more and more prominent. To take care of different plan issues, it is seen that structure design finds a superior answer for a large number of repetitive plan issues. For the most part, design-level documents are indicated utilising semi-formal documentation, for example, unified modelling language (UML) diagrams. In any case, these kinds of semi-formal documentation lead to ambiguities and irregularities. In this paper, design level documents are retrieved by using the idea of program metrics. The proposed methodology retrieves the reusable documents from the open-source software such as PMD and JUnit. The presented method uses three kinds of neural network models to analyse the effectiveness of the pattern retrieval approach.

Dynamic pattern matching with multiple queries on large scale data streams

Similarity search in data streams is an important but challenging task in many practical areas where real-time pattern retrieval is required. Dynamic and fast updating data streams are often subject to outliers, noise and potential distortions in amplitude and time dimensions. Such conditions typically lead to a failure of existing pattern matching algorithms and to inability to retrieve required patterns from the stream. The main reason for such failures is the limitation of data normalization utilized in the majority of methods. Another reason is the lack of means to consider multiple examples of the same template to account for possible variations of the query signal. In this paper, we propose a dynamic normalization approach that allows bringing streaming signal subsequences to the scale of the query template. This significantly improves pattern retrieval capabilities, especially when sampling variance or time distortions are present. We further develop a pattern matching approach utilizing the proposed normalization mechanism and extend it for the case when multiple examples of a query template are available. Multiple synthetic and real data experiments demonstrate that this allows to considerably improve the pattern matching rate for distorted data streams, providing real time performance.

Data integrity management for detection of redundancy and recurrence patterns in cloud

Cloud computing and larger data storage unit’s expansions has become a major research challenge amongst researchers. The expanding resources with respect to data is practically nor a solution towards archiving data and resources. In cloud computing environment the data is replicated and recursively stored. Thus the proposed methodology defines a solution towards data optimization and indexing. An index based searching technique for data collection and similarity analysis. Similarity analysis is processed with retrieval of data patterns under defined data attributed. Thus, the system to be optimized in storage. The experimental setup is conducted on local connected servers of primary specific organization. The results demonstrate 43.2% optimization of given space.

Retrieval of spatial representation on network level in hippocampal CA3 accompanied by overexpression and mixture of stored network patterns

Retrieval of stored network activity pattern has been shown as a competitive transition from one attractor state to another, orchestrated by local theta oscillation. However, the fine nature of this process that is considered as substrate of memory recall is not clear. We found that hippocampal network recall is characterized by hyperactivity in the CA3 place cell population, associated with an “overexpression” of the retrieved network pattern. The overexpression was based on recruitment of cells from the same (recalled) spatial representation with low expected firing probability at the given position. We propose that increased place cell activation during state transitions might facilitate pattern completion towards the retrieved network state and stabilize its expression in the network. Furthermore, we observed frequent mixing of both activity patterns at the temporal level of a single theta cycle. On a sub-theta cycle scale, we found signs of segregation that might correspond to a gamma oscillation patterning, as well as occasional mixing at intervals of less than 5 milliseconds. Such short timescale coactivity might induce plasticity mechanisms, leading to associations across the two originally decorrelated network activity states.

Физический эксперимент по реконструкции диаграммы направленности антенны в рамках методологии сходящегося пучка плоских волн.

Физический эксперимент по реконструкции диаграммы направленности антенны в рамках методологии сходящегося пучка плоских волн.

Targeted Memory Reactivation during Sleep Elicits Neural Signals Related to Learning Content.

Retrieval of learning-related neural activity patterns is thought to drive memory stabilization. However, finding reliable, noninvasive, content-specific indicators of memory retrieval remains a central challenge. Here, we attempted to decode the content of retrieved memories in the EEG during sleep. During encoding, male and female human subjects learned to associate spatial locations of visual objects with left- or right-hand movements, and each object was accompanied by an inherently related sound. During subsequent slow-wave sleep within an afternoon nap, we presented half of the sound cues that were associated (during wake) with left- and right-hand movements before bringing subjects back for a final postnap test. We trained a classifier on sleep EEG data (focusing on lateralized EEG features that discriminated left- vs right-sided trials during wake) to predict learning content when we cued the memories during sleep. Discrimination performance was significantly above chance and predicted subsequent memory, supporting the idea that retrieval leads to memory stabilization. Moreover, these lateralized signals increased with postcue sleep spindle power, demonstrating that retrieval has a strong relationship with spindles. These results show that lateralized activity related to individual memories can be decoded from sleep EEG, providing an effective indicator of offline retrieval.SIGNIFICANCE STATEMENT Memories are thought to be retrieved during sleep, leading to their long-term stabilization. However, there has been relatively little work in humans linking neural measures of retrieval of individual memories during sleep to subsequent memory performance. This work leverages the prominent electrophysiological signal triggered by lateralized movements to robustly demonstrate the retrieval of specific cued memories during sleep. Moreover, these signals predict subsequent memory and are correlated with sleep spindles, neural oscillations that have previously been implicated in memory stabilization. Together, these findings link memory retrieval to stabilization and provide a powerful tool for investigating memory in a wide range of learning contexts and human populations.

Quantum accelerated approach to the thermal state of classical all-to-all connected spin systems with applications to pattern retrieval in the Hopfield neural network

We explore the question as to whether quantum effects can yield a speedup of the non-equilibrium evolution of spin systems towards a classical thermal state. In our approach we exploit the fact that the thermal state of a spin system can be mapped onto a node-free quantum state whose coefficients are given by thermal weights. This perspective permits the construction of a dissipative -- yet quantum -- dynamics which encodes in its stationary state the thermal state of the original problem. We show for the case of an all-to-all connected Ising spin model that an appropriate transformation of this dissipative dynamics allows to interpolate between a regime in which the order parameter obeys the classical equations of motion under Glauber dynamics, to a quantum regime with an accelerated approach to stationarity. We show that this effect enables in principle a speedup of pattern retrieval in a Hopfield neural network.

Efficient phase retrieval of two-directional phase-shifting fringe patterns using geometric constraints of deflectometry.

In the phase measuring deflectometry, two groups of fringe patterns in orthogonal directions are usually applied to establish the correspondences between the pixel pairs on the screen and camera. Usually, 16 phase-shifting fringe patterns with different spatial frequencies are required in order to calculate the absolute phases in the conventional temporal phase unwrapping algorithms. This requirement makes the measurement inefficient and not robust against environmental noise. In this paper, an efficient phase retrieval strategy is developed, which requires only six fringe patterns. The modulating information in one-direction is obtained by first using four fringe patterns, and then it is applied to assist the phase calculation in the other direction, so that only two extra fringe patterns are needed. Subsequently the phases are unwrapped by using the geometric constraints of the software configurable optical test system without additional image acquisition. The measurement time is saved by 5/8, compared to the conventional methods. In this way, the influence of the low-frequency disturbances can be suppressed in the workshop condition. Experiments demonstrate that the proposed method can reliably retrieve the absolute phases, and it is of significance to improve the measuring efficiency and stability of in situ deflectometry.

Domain-specific data mining for residents' transit pattern retrieval from incomplete information

The rapid development of Cyber-Physical System (CPS) is gradually cultivating a smart world based on Big Data and computational infrastructures. Such reformation has initiated revolutions in various industries and is gradually reshaping our daily routines. Among these trends, the data-driven intelligentization in Urban Public Transportation Systems can bring the most significant impact to our society, because they provide pervasive reachability. Optimizing and constructing such systems, which requires deep insights into passenger behavioral patterns, is a highly domain-specific data mining task. Although computational infrastructures have provided sufficient processing capacity, the practical utilization of such data is facing several challenges: information alignment in heterogeneous data sources and information enrichment for domain-specific applications. In this article, we first propose a cross-domain method to increase the usability of data by removing the inconsistency in vehicles' positioning and passengers' transaction data. We then use rule-based methods to reconstruct latent mobility information, thereby enabling small grained trajectory based applications. Finally, we provide use cases using the reconstructed information to derive insights to the public transportation system of our target city. Our work can serve as a domain-specific information enrichment and data mining framework for CPS in smart cities.

Direct three-dimensional tomographic reconstruction and phase retrieval of far-field coherent diffraction patterns

We present an alternative numerical reconstruction algorithm for direct tomographic reconstruction of a sample refractive indices from the measured intensities of its far-field coherent diffraction patterns. We formulate the well-known phase-retrieval problem in ptychography in a tomographic framework which allows for simultaneous reconstruction of the illumination function and the sample refractive indices in three dimensions. Our iterative reconstruction algorithm is based on the Levenberg-Marquardt algorithm. We demonstrate the performance of our proposed method with simulation studies.

Hope4Genes: a Hopfield-like class prediction algorithm for transcriptomic data

After its introduction in 1982, the Hopfield model has been extensively applied for classification and pattern recognition. Recently, its great potential in gene expression patterns retrieval has also been shown. Following this line, we develop Hope4Genes a single-sample class prediction algorithm based on a Hopfield-like model. Differently from previous works, we here tested the performances of the algorithm for class prediction, a task of fundamental importance for precision medicine and therapeutic decision-making. Hope4Genes proved better performances than the state-of-art methodologies in the field independently of the size of the input dataset, its profiling platform, the number of classes and the typical class-imbalance present in biological data. Our results provide encoraging evidence that the Hopfield model, together with the use of its energy for the estimation of the false discoveries, is a particularly promising tool for precision medicine.

Retrieving color patterns on surface meshes using edgeLBP descriptors

Texture image retrieval and classification tasks are conducted widely in computer vision but few studies have considered the same problem for decorations and textures applied to surfaces. Thus, in order to address this problem, we propose an algorithm for retrieving and classifying color patterns over surface meshes. In this study, we explored the capability of edgeLBP descriptors for retrieving and classifying color patterns over surface tessellations. Our experimental results demonstrated that these descriptors performed well at pattern retrieval, including for patterns on degraded and corrupted archeological fragments. We compared the performance of the edgeLBP descriptors with a recent benchmark and conducted an extended analysis of the robustness of this technique to mesh remeshing and surface decimation.

Retrieval of Similar Evolution Patterns from Satellite Image Time Series

Technological evolution in the remote sensing domain has allowed the acquisition of large archives of satellite image time series (SITS) for Earth Observation. In this context, the need to interpret Earth Observation image time series is continuously increasing and the extraction of information from these archives has become difficult without adequate tools. In this paper, we propose a fast and effective two-step technique for the retrieval of spatio-temporal patterns that are similar to a given query. The method is based on a query-by-example procedure whose inputs are evolution patterns provided by the end-user and outputs are other similar spatio-temporal patterns. The comparison between the temporal sequences and the queries is performed using the Dynamic Time Warping alignment method, whereas the separation between similar and non-similar patterns is determined via Expectation-Maximization. The experiments, which are assessed on both short and long SITS, prove the effectiveness of the proposed SITS retrieval method for different application scenarios. For the short SITS, we considered two application scenarios, namely the construction of two accumulation lakes and flooding caused by heavy rain. For the long SITS, we used a database formed of 88 Landsat images, and we showed that the proposed method is able to retrieve similar patterns of land cover and land use.

Frequent Pattern Retrieval on Data Streams by using Sliding Window

In different applications like recommender frameworks and market examination, regular patterns play a significant role in useful mining data. Mining regular patterns from sliding windows over streaming information has become a complex task. In this examination, the sliding window is utilized