We consider here the task of multi-label classification for data organized in a multi-relational graph. We propose the IMMCA model - iterative Multi-label Multi-Relational Classification algorithm - a general algorith...
详细信息
We consider here the task of multi-label classification for data organized in a multi-relational graph. We propose the IMMCA model - iterative Multi-label Multi-Relational Classification algorithm - a general algorithm for solving the inference and learning problems for this task. Inference is performed iteratively by propagating scores according to the multi-relational structure of the data. We detail two instances of this general model, implementing two different label propagation schemes on the multi-graph. This is the first collective classification method able to handle multiple relations and to perform multi-label classification in multi-graphs. The target application is image annotation in large social media sharing web sites (Flickr). The goal is to assign labels for images when users and images are connected through multiple relations - authorship, friendship, or visual/textual similarities. We show that our model is able to deal with both content and social relations and performs well on real datasets. Additional experiments on artificial data allow us analyzing the behavior of our method in different situations.
This paper presents an iterative learning control scheme for a class of uncertain systems in the presence of initial state errors. The terminal attractor is adopted in the control design. An iterative learning algorit...
详细信息
This paper presents an iterative learning control scheme for a class of uncertain systems in the presence of initial state errors. The terminal attractor is adopted in the control design. An iterative learning algorithm is proposed to deal with time-varying parametric uncertainties, while a periodic learning algorithm is given to cope with time-invariant ones. It is shown through the theoretical analysis and numerical simulation that system states completely follow the desired trajectories after the pre-specified time, and all the signals in the closed-loop are guaranteed to be bounded.
This paper presents the implementation of a crossbar-based CIOQ switch with a novel matching algorithm that can be used for efficient interprocessor communication. The proposed matching algorithm, called SIM, uses a s...
详细信息
This paper presents the implementation of a crossbar-based CIOQ switch with a novel matching algorithm that can be used for efficient interprocessor communication. The proposed matching algorithm, called SIM, uses a sequential mode to replace the traditional parallel mode to improve the efficiency of iterative matching operation. Moreover, the paper presents a FIFO-based request scheduler for the implementation of SIM. The proposed algorithm and its implementation architecture are evaluated in an FPGA. The implementation-based simulation results show that for switches without speedup, SIM can achieve higher aggregate throughput and lower average delay than iSLIP. Moreover, SIM achieves a fair trade-off among performance and logic area.
Modern deconvolution algorithms are often specified as minimization problems involving a non-quadratic regularization functional. When the latter is a wavelet-domain ℓ 1 -norm that favors sparse solutions, the problem...
详细信息
Modern deconvolution algorithms are often specified as minimization problems involving a non-quadratic regularization functional. When the latter is a wavelet-domain ℓ 1 -norm that favors sparse solutions, the problem can be solved by a simple iterative shrinkage/thresholding algorithm (ISTA). This approach provides state-of-the-art results in 2-D, but is harder to deploy in 3-D because of its slow convergence. In this paper, we propose an acceleration scheme that turns wavelet-regularized deconvolution into a competitive solution for 3-D fluorescence microscopy. A significant speed-up is achieved though a synergistic combination of subband-adapted thresholds and sequential TwIST updates. We provide a theoretical justification of the procedure together with an experimental evaluation, including the application to real 3-D fluorescence data.
In this paper, we propose a linear transfer operator-based approach for the computation of domain of attraction of an attractor set. The almost everywhere notion of stability of an attractor set is used for the comput...
详细信息
ISBN:
(纸本)9781424477456
In this paper, we propose a linear transfer operator-based approach for the computation of domain of attraction of an attractor set. The almost everywhere notion of stability of an attractor set is used for the computation of domain of attraction. The proposed method is applicable to autonomous dynamical systems with the only requirement that the system mapping is a differentiable function of states. Furthermore domain of attraction of any general attractor set can be computed. Set oriented numerical scheme and iterative procedure based on the solution of finite dimensional linear program is proposed for the computation of domain of attraction. The application of the proposed method is demonstrated on two and three dimensional continuous and discrete-time systems.
This paper presents a novel Web service composition algorithm based on TOPSIS (WSC_TOPSIS) to solve the service composition difficulties with multiple decision-makers and heterogeneous QoS for the first time. It inclu...
详细信息
This paper presents a novel Web service composition algorithm based on TOPSIS (WSC_TOPSIS) to solve the service composition difficulties with multiple decision-makers and heterogeneous QoS for the first time. It includes three main steps: normalizing decision matrix, evaluating alternatives synthetically and evaluating group alternatives synthetically. Experimental results show that the proposed algorithm can better support Web service composition with heterogeneous QoS data and multiple decision-makers.
The paper presents a novel algorithm for multiple space vehicles *** algorithm first positions the ground stations using multiple space vehicles with initial coordinates, then calculates the differences of the ground ...
详细信息
The paper presents a novel algorithm for multiple space vehicles *** algorithm first positions the ground stations using multiple space vehicles with initial coordinates, then calculates the differences of the ground stations position results and their real coordinates,at last uses the differences to estimate the revised error for space vehicles’*** can use iterative method to achieve needed precision of the SV’s *** last,simulation results prove the stability and credibility of the algorithm in space vehicles positioning.
Some shortest path problems that can be solved using the A* algorithm have a large branching factor due to the combination of multiple choices at each move. Multiple sequence alignment and multi-agent pathfinding are ...
详细信息
Some shortest path problems that can be solved using the A* algorithm have a large branching factor due to the combination of multiple choices at each move. Multiple sequence alignment and multi-agent pathfinding are examples of such problems. If the search can be stopped after each choice instead of being stopped at each combination of choices, it takes much less memory and much less time. The goal of the paper is to show that Partial Move A* is much better than A* for these problems when the branching factor is large due to a large combination of choices at each move. When there is such a large combination of choices at each move, Partial Move A* can yield large memory gains and speedups over regular A*.
The OFDM systems with the serial concatenated convolutional codes(SCCC)interleave the information random in the time field(COFDM).It is able to overcome the abrupt noises and enhance the performance of the whole syste...
详细信息
The OFDM systems with the serial concatenated convolutional codes(SCCC)interleave the information random in the time field(COFDM).It is able to overcome the abrupt noises and enhance the performance of the whole system *** in the actual communication systems,the frequency offset deteriorates the performance of COFDM *** this problem becomes more complex,because the frequency offset is relative to the difference of frequency between transmitter and receiver and the Doppler shifts of channel those are *** paper combines the MAP decoding algorithm of SCCC with the expectation-maximization(EM)algorithm of tracking frequency offset(MAP-EM),and uses the feedback iterative technique to implement the estimate of both the frequency offset and *** technique can enhance the performance of systems,because it bases on the ML and the information could be helped each other between the OFDM *** simulations show that the more iterative the better performance of the tracking precision and the BER of whole systems.
Crosstalk is the major limiting issue in VDSL2 system. Based on matrix splitting and iterative theory, a general form of iterative crosstalk cancellation algorithm is first proposed for VDSL2 upstream transmission. Fr...
详细信息
ISBN:
(纸本)9781424448562
Crosstalk is the major limiting issue in VDSL2 system. Based on matrix splitting and iterative theory, a general form of iterative crosstalk cancellation algorithm is first proposed for VDSL2 upstream transmission. From using diagonal dominant and column wise diagonal dominant of crosstalk channel matrix property, two special iterative algorithms are then derived. Computer simulation results based on measured data verify that the iterative crosstalk cancellation algorithms can improve the performance and approach zero-forcing algorithm by successive iterations.
暂无评论