Machine Representation Research Articles

Multiple data-dependent kernel for classification of hyperspectral images

Kernel-based learning strategies have recently emerged as powerful tools for hyperspectral classification. However, designing optimal kernels is still a challenging issue that needs to be further investigated. In this paper, we propose a multiple data-dependent kernel (MDK) for classification of HSI. Core ideas of the MDK are twofold: (1) optimizing the combination of multiple basic kernels in merit of centered kernel alignment (CKA), which can evaluate the degree of agreement between a kernel and a learning task; (2) optimizing the coefficients of data-dependent kernel (DK) by virtue of Fisher’s discriminant analysis (FDA), which can measure the between-class and within-class separability of the data simultaneously. Furthermore, we apply the proposed MDK to two state-of-the-art classifiers, i.e. support vector machine (SVM) and sparse representation classifier (SRC). Experimental results conducted on three benchmark HSIs with different spectral and spatial resolutions validate the feasibility of the proposed methods.

Efficiently Implementing GOLOG with Answer Set Programming

In this paper we investigate three different approaches to encoding domain-dependent control knowledge for Answer-Set Planning. Starting with a standard imple- mentation of the action description language B, we add control knowledge expressed in the GOLOG logic pro- gramming language. A naive encoding, following the original definitions of Levesque et al., is shown to scale poorly. We examine two alternative codings based on the transition semantics of ConGOLOG. We show that a speed increase of multiple orders of magnitude can be obtain by compiling the GOLOG program into a finite- state machine representation.

A fuzzy ontology for semantic modelling and recognition of human behaviour

We propose a fuzzy ontology for human activity representation, which allows us to model and reason about vague, incomplete, and uncertain knowledge. Some relevant subdomains found to be missing in previous proposed ontologies for this domain were modelled as well. The resulting fuzzy OWL 2 ontology is able to model uncertain knowledge and represent temporal relationships between activities using an underlying fuzzy state machine representation. We provide a proof of concept of the approach in work scenarios such as the office domain, and also make experiments to emphasize the benefits of our approach with respect to crisp ontologies. As a result, we demonstrate that the inclusion of fuzzy concepts and relations in the ontology provide benefits during the recognition process with respect to crisp approaches.

Optimizing Spatial and Temporal Reuse in Wireless Networks by Decentralized Partially Observable Markov Decision Processes

The performance of medium access control (MAC) depends on both spatial locations and traffic patterns of wireless agents. In contrast to conventional MAC policies, we propose a MAC solution that adapts to the prevailing spatial and temporal opportunities. The proposed solution is based on a decentralized partially observable Markov decision process (DEC-POMDP), which is able to handle wireless network dynamics described by a Markov model. A DEC-POMDP takes both sensor noise and partial observations into account, and yields MAC policies that are optimal for the network dynamics model. The DEC-POMDP MAC policies can be optimized for a freely chosen goal, such as maximal throughput or minimal latency, with the same algorithm. We make approximate optimization efficient by exploiting problem structure: the policies are optimized by a factored DEC-POMDP method, yielding highly compact state machine representations for MAC policies. Experiments show that our approach yields higher throughput and lower latency than CSMA/CA based comparison methods adapted to the current wireless network configuration.

비디오 행동 인식을 위하여 다중 판별 결과 융합을 통한 성능 개선에 관한 연구

최근 다양한 방송 및 영상 분야에서 사람의 행동을 인식하여는 연구들이 많이 이루어지고 있다. 영상은 다양한 형태를 가질 수 있기 때문에 제약된 환경에서 유용한 템플릿 방법들보다 특징점에 기반한 연구들이 실제 사용자 환경에서 더욱 관심을 받고 있다. 특징점 기반의 연구들은 영상에서 움직임이 발생하는 지점들을 찾아내어 이를 3차원 패치들로 생성한다. 이를 이용하여 영상의 움직임을 히스토그램에 기반한 descriptor(서술자)로 표현하고 학습기반의 판별기로 최종적으로 영상내에 존재하는 행동들을 인식하였다. 그러나 단일 판별기로는 다양한 행동을 인식하기에 어려움이 있다. 따라서 이러한 문제를 개선하기 위하여 최근에 다중 판별기를 활용한 연구들이 영상 판별 및 물체 검출 영역에서 사용되고 있다. 따라서 본 논문에서는 행동 인식을 위하여 support vector machine과 sparse representation을 이용한 decision-level fusion 방법을 제안하고자 한다. 제안된 논문의 방법은 영상에서 특징점 기반의 descriptor를 추출하고 이를 각각의 판별기를 통하여 판별 결과들을 획득한다. 이 후 학습단계에서 획득된 가중치를 활용하여 각 결과들을 융합하여 최종 결과를 도출하였다. 본 논문에 실험에서 제안된 방법은 기존의 융합 방법보다 높은 행동 인식 성능을 보여 주었다. Recently, human action recognition have been developed for various broadcasting and video process. Since a video can consist of various scenes, keypoint approaches have been more attracted than template based methods for real application. Keypoint approahces tried to find regions having motion in video, and made 3-dimensional patches. Then, descriptors using histograms were computed from the patches, and a classifier based on machine learning method was applied to detect actions in video. However, a single classifier was difficult to handle various human actions. In order to improve this problem, approaches using multi classifiers were used to detect and to recognize objects. Thus, we propose a new human action recognition using decision-level fusion with support vector machine and sparse representation. The proposed method extracted descriptors based on keypoint approach from a video, and acquired results from each classifier for human action recognition. Then, we applied weights which were acquired by training stage to fuse each results from two classifiers. The experiment results in this paper show better result than a previous fusion method.

Karhunen-Loeve Transform and Sparse Representation Based Plant Leaf Disease Recognition

To improve the classification accuracy rate of apple leaf disease images and solve the problem of dimension redundancy in feature extraction, Karhunen-Loeve (K-L) transform and sparse representation are applied to apple leaf disease recognition. Firstly 9 color features and 8 texture features of disease leaf images are extracted and taken as feature vectors after dimensionality reduction by the K-L transform. Then, for each of apple mosaic virus, apple rust and apple alternaria leaf spot, 40 apple leaf images are selected as the training samples, whose feature vectors are made up of the dictionary of the sparse representation, respectively. Each testing sample is classified into the class with the minimal residual. The identifying results using the proposed method are analyzed and compared with those of the Support Vector Machine (SVM) and original sparse representation method. The average classification accuracy rate of the proposed method is 94.18 %, which confirms its good robustness. In addition, the proposed method not only improves the plant leaf disease classification accuracy but also solves the redundancy problem of the extracted features. DOI : http://dx.doi.org/10.11591/telkomnika.v12i2.3632

Automated State Machine and Timing Characteristic Extraction for RSFQ Circuits

We present a heuristic for the automated extraction of state machine representations of rapid single-flux-quantum (RSFQ) digital logic circuits given a circuit SPICE representation. Furthermore, this heuristic uses the SPICE netlist to extract timing characteristics for the creation of a hardware description language (HDL) implementation of the circuit. This facilitates RSFQ logic design at an HDL level rather than a Josephson junction level. The state machine extraction method can be also used for the automatic creation of test benches for circuit yield analysis, as well as optimization algorithms. An example of the automatic extraction of the state machine representation, timing characterization, and HDL implementation is demonstrated for a complex RSFQ cell.

Place reference systems

Reference to places is a central but largely underexposed problem of information science. Place has been a major object of research in many domains including Geography, Cognitive Science and Geographic Information Science. However, Geographic Information Systems (GIS) have been built solely on space reference systems creating a gap between human conceptualization and machine representation. While reference to space only partially captures reference to place, most existing definitions of place either reduce the latter to the former or lack a formal characterization of how places are constructed. In a spatial coordinate system, locations are referenced by angles and distances to other referents. In this paper, we suggest that place reference systems can be built based on localizing things (locatums) involved in simulated activities relative to other involved referents (locators). We propose a formal theory about relevant types of activities and their involved participants, and show how place referents can be identified and localized by choosing locators and locatum among the participants. We formally derive an ontology of places, publish a corresponding OWL version, and demonstrate how to compute a market place and a vantage place in a GIS.

Change detection in precision manufacturing processes under transient conditions

Early detection of changes in transient process behaviors from sensor signals is becoming essential for quality assurance in microelectronics and ultraprecision manufacturing processes. We present a Dirichlet process Gaussian State Machine (DPGSM) representation to capture complex dynamics as a random concatenation of nonlinear stationary segments, and develop a method to detect early-stage fault-inducing changes. Extensive experiments suggest that the present approach, compared to other methods tested, was able to detect slight changes that cause severe surface damage 48ms earlier in an ultraprecision machining (UPM) process, and at least 2000ms earlier in a chemical mechanical planarization (CMP) process.

Efficient Localization of Discontinuities in Complex Computational Simulations

Surrogate models for computational simulations are input-output approximations that allow computationally intensive analyses, such as uncertainty propagation and inference, to be performed efficiently. When a simulation output does not depend smoothly on its inputs, the error and convergence rate of many approximation methods deteriorate substantially. This paper details a method for efficiently localizing discontinuities in the input parameter domain, so that the model output can be approximated as a piecewise smooth function. The approach comprises an initialization phase, which uses polynomial annihilation to assign function values to different regions and thus seed an automated labeling procedure, followed by a refinement phase that adaptively updates a kernel support vector machine representation of the separating surface via active learning. The overall approach avoids structured grids and exploits any available simplicity in the geometry of the separating surface, thus reducing the number of model ...

On-the-fly detection of instability problems in floating-point program execution

The machine representation of floating point values has limited precision such that errors may be introduced during execution. These errors may get propagated and magnified by the following operations, leading to instability problems, e.g., control flow path may be undesirably altered and faulty output may be emitted. In this paper, we develop an on-the-fly efficient monitoring technique that can predict if an execution is stable. The technique does not explicitly compute errors as doing so incurs high overhead. Instead, it detects possible places where an error becomes substantially inflated regarding the corresponding value, and then tags the value with one bit to denote that it has an inflated error. It then tracks inflation bit propagation, taking care of operations that may cut off such propagation. It reports instability if any inflation bit reaches a critical execution point, such as a predicate, where the inflated error may induce substantial execution difference, such as different execution paths. Our experiment shows that with appropriate thresholds, the technique can correctly detect that over 99.999996% of the inputs of all the programs we studied are stable while a traditional technique relying solely on inflation detection mistakenly classifies majority of the inputs as unstable for some of the programs. Compared to the state of the art technique that is based on high precision computation and causes several hundred times slowdown, our technique only causes 7.91 times slowdown on average and can report all the true unstable executions with the appropriate thresholds.

Validation of a Faulted Rotor Induction Machine Model With an Insightful Geometrical Interpretation of Physical Quantities

Many proposals about induction-machine broken-bar detection can be found in the literature. Disturbances that affect phase currents, current space vector, power, fluxes, torque, and speed have been utilized to diagnose mains-supplied machines. Nevertheless, there is still disagreement about how the fault signatures of the different variables correlate with the actual fault and how they depend on the operating conditions. This paper presents a simple but effective analytical representation of induction machines with rotor faults, based on a synchronous reference frame and on Steinmetz's phasors. The developed relationships can be visualized through equivalent circuits, feedback loop block diagrams, or geometrical loci, allowing a deep understanding of the underlying physical phenomena, particularly as functions of inertia and of the machine parameters. The insight thus gained helps to enhance the quantification of fault severity with respect to the usual diagnostic technique based on the sum of the sidebands, clearly defining for the first time the field of validity of such technique. The analytical results are verified through simulations and experiments.

Identifying the effects of modifications as data dependencies

Dependence analysis on an extended finite state machine (EFSM) representation of the requirements of a system under test has been used in model-based regression testing for regression test suite (RTS) reduction (reducing the size of a given test suite by eliminating redundancies), for RTS prioritization (ordering test cases in a given test suite for early fault detection), or for RTS selection (selecting a subset of a test suite covering the identified dependencies). These particular uses of dependence analysis are based on the definitions of various types of control and data dependencies (between transitions in an EFSM) caused by a given set of modifications on the requirements. This paper considers the existing definitions of data dependencies for capturing the effects of the modifications, gives examples of their inaccuracy and incompleteness, proposes new definitions, and proves the soundness and completeness of these new definitions. Any previous work on regression testing using definitions of data dependencies capturing the effects of modifications can benefit from the proposed definitions.

A hybrid approach combining extreme learning machine and sparse representation for image classification

Two well-known techniques, extreme learning machine (ELM) and sparse representation based classification (SRC) method, have attracted significant attention due to their respective performance characteristics in computer vision and pattern recognition. In general, ELM has speed advantage and SRC has accuracy advantage. However, there also remain drawbacks that limit their practical application. Actually, in the field of image classification, ELM performs extremely fast while it cannot handle noise well, whereas SRC shows notable robustness to noise while it suffers high computational cost. In order to incorporate their respective advantages and also overcome their respective drawbacks, this work proposes a novel hybrid approach combining ELM and SRC for image classification. The new approach is applied to handwritten digit classification and face recognition, experiments results demonstrate that it not only outperforms ELM in classification accuracy but also has much less computational complexity than SRC.

Representing Objects, Relations, and Sequences

Vector symbolic architectures (VSAs) are high-dimensional vector representations of objects (e.g., words, image parts), relations (e.g., sentence structures), and sequences for use with machine learning algorithms. They consist of a vector addition operator for representing a collection of unordered objects, a binding operator for associating groups of objects, and a methodology for encoding complex structures. We first develop constraints that machine learning imposes on VSAs; for example, similar structures must be represented by similar vectors. The constraints suggest that current VSAs should represent phrases ("The smart Brazilian girl") by binding sums of terms, in addition to simply binding the terms directly. We show that matrix multiplication can be used as the binding operator for a VSA, and that matrix elements can be chosen at random. A consequence for living systems is that binding is mathematically possible without the need to specify, in advance, precise neuron-to-neuron connection properties for large numbers of synapses. A VSA that incorporates these ideas, Matrix Binding of Additive Terms (MBAT), is described that satisfies all constraints. With respect to machine learning, for some types of problems appropriate VSA representations permit us to prove learnability rather than relying on simulations. We also propose dividing machine (and neural) learning and representation into three stages, with differing roles for learning in each stage. For neural modeling, we give representational reasons for nervous systems to have many recurrent connections, as well as for the importance of phrases in language processing. Sizing simulations and analyses suggest that VSAs in general, and MBAT in particular, are ready for real-world applications.

SYSTEM PERSONALITY AND ADAPTIVITY IN AFFECTIVE HUMAN-COMPUTER INTERACTION

It has been demonstrated that human users attribute a personality to the computer interfaces they use, regardless of whether one has been explicitly encoded in the system's design or not. In this paper, we explore a method for having explicit control over the personality that a spoken human-robot interface is perceived to exhibit by its users. Our method focuses on the interaction between users and semantic knowledge-based systems where the goal of the interaction is that information from the semantic store is relayed to the user. We describe a personality modelling method that complements a standard dialogue manager by calculating parameters related to adaptivity and emotion for the various interaction modules that realize the system's dialogue acts. This calculation involves the planned act, the user adaptivity model, the system's own goals, but also a machine representation of the personality that we want the system to exhibit, so that systems with different personality will react differently even when in the same dialogue state and with the same user or user type.

Improved Sequence Network Model of Wind Turbine Generators for Short-Circuit Studies

Protective relay settings for wind turbine generators often depend on short-circuit calculations performed using sequence network circuits in rms-based protection software. Traditional assumptions made in deriving the sequence network representation of induction machines result in some error in short-circuit calculations, which could potentially cause incorrect relay settings. A more accurate sequence network model of induction machines is derived in this paper to aid in short-circuit calculations in induction-generator-based wind plants. A rigorous mathematical approach to defining the stator short-circuit current equations under general faults is described and validated with transient simulations. The sequence network model of the induction machine is derived from the closed-form mathematical solutions. An example short-circuit calculation is performed on a simple network in this paper, and found to have a marked improvement in accuracy from traditional sequence network calculation methods for unbalanced faults. Resonant effects of power factor correction capacitors are also described.

Type Classes of Context Trees

It is well known that a tree model does not always admit a finite-state machine (FSM) representation with the same (minimal) number of parameters. Therefore, known characterizations of type classes for FSMs do not apply, in general, to tree models. In this paper, the type class of a sequence with respect to a given context tree is studied. An exact formula is derived for the size of the class, extending Whittle's formula for type classes with respect to FSMs. The derivation is more intricate than in the FSM case, since some basic properties of FSM types do not hold in general for tree types. The derivation also yields an efficient enumeration of the tree type class. A formula for the number of type classes with respect to is also derived. The formula is asymptotically tight up to a multiplicative constant and also extends the corresponding result for FSMs. The asymptotic behavior of the number of type classes, and of the size of a class, is expressed in terms of the so-called minimal canonical extension of T, a tree that is generally larger than but smaller than its FSM closure.

Regression test suite selection using dependence analysis

SUMMARYDependence analysis on an Extended Finite State Machine representation of the requirements of a system under test identifies various types of control and data dependencies between transitions caused by a set of modifications on the requirements. These particular types of dependencies capture the effects of the modifications, that is, their direct effects on the changed parts of the system and their side effects on the unchanged parts of the system. Recent work on model‐based regression testing shows that dependencies capturing direct effects and side effects of the changes made on the requirements can be used for regression test suite (RTS) reduction (reducing the size of a given test suite by eliminating redundancies), for RTS prioritization (ordering test cases in a given test suite for early fault detection), or for RTS generation (designing a test suite covering the identified dependencies). This paper proposes an additional use of such dependencies, namely, RTS selection, which is the process of selecting a subset of a given test suite to form an RTS by considering the coverage of dependencies related to the effects of the modifications. The dependencies marked during this process as uncovered provide a basis for augmenting an (incomplete) RTS with test cases covering uncovered dependencies. Copyright © 2012 John Wiley & Sons, Ltd.

Impact of the Use of Object Request Broker Middleware for Inter-Component Communications in C6416 Digital Signal Processor Based Software Communications Architecture Radio Systems

Problem statement: This study presents an in-depth analysis of the performance of Software Communications Architecture (SCA) component-based waveform applications in terms of inter-component communications. The main limitation with SCA, in the context of embedded systems, is the additional cost introduced by the use of Object Request Broker (ORB) middleware. The ORB middleware handles the interaction between components and objects in SCA distributed environment. This interaction should be highly efficient, due to the real time nature of SCA systems and transparent to the application programmer. Approach: We can achieve high efficiency in SCA systems by enhancing the Inter-Process Communications (IPC) mechanisms in Operating systems (OS) micro kernels, while we achieve transparency through Interface Definition Language (IDL). Different encoding mechanisms like “External Data Representation (XDR), Network Data Representation (NDR) and Common Data Representation (CDR) facilitate inter-component communication transparently and efficiently”. Marshalling procedures format data from the local machine representation to common network representations. A most common encoding mechanism for Common Object Request Broker Architecture (CORBA) systems is CDR representation. Measurements have been performed with ORBExpress DSP as a CORBA distribution and Open Source SCA Implementation Embedded (OSSIE) for SCA implementation. In order to perform these measurements we proposed two metrics for profiling the ORB that are invocation and marshalling. In addition, we propose three elements of data types to evaluate the performance of ORB middleware that are, Basic, Array and Sequence data types. Results: The CORBA bus is really the part, which brings an overhead to the SCA radio systems. This overhead is due to method invocations that have been carried out by ORB middleware. Conclusion: Performance benchmarks of ORBExpress DSP middleware show that, although using CORBA for inter-component communications introduces delays and overheads, the overall effect can be reduced by sending packets of data instead of basic data type elements.