Ensemble learning has many successful applications because of its effectiveness in boosting the predictive performance of classification models. In this article, we propose a semisupervised multiple choice learning (SemiMCL) approach to jointly train a network ensemble on partially labeled data. Our model mainly focuses on improving a labeled data assignment among the constituent networks and exploiting unlabeled data to capture domain-specific information, such that semisupervised classification can be effectively facilitated. Different from conventional multiple choice learning models, the constituent networks learn multiple tasks in the training process. Specifically, an auxiliary reconstruction task is included to learn domain-specific representation. For the purpose of performing implicit labeling on reliable unlabeled samples, we adopt a negative$\ell _{1}$-norm regularization when minimizing the conditional entropy with respect to the posterior probability distribution. Extensive experiments on multiple real-world datasets are conducted to verify the effectiveness and superiority of the proposed SemiMCL model.

Semi-supervised generative learning (SSGL) makes use of unlabeled data to achieve a trade-off between the data collection/annotation effort and generation performance, when adequate labeled data are not available. Learning precise class semantics is crucial for class-conditional image synthesis with limited supervision. Toward this end, we propose a semi-supervised Generative Adversarial Network with a Mask-Embedded Discriminator, which is referred to as MED-GAN. By incorporating a mask embedding module, the discriminator features are associated with spatial information, such that the focus of the discriminator can be limited in the specified regions when distinguishing between real and synthesized images. A generator is enforced to synthesize the instances holding more precise class semantics in order to deceive the enhanced discriminator. Also benefiting from mask embedding, region-based semantic regularization is imposed on the discriminator feature space, and the degree of separation between real and fake classes and among object categories can thus be increased. This eventually improves class-conditional distribution matching between real and synthesized data. In the experiments, the superior performance of MED-GAN demonstrates the effectiveness of mask embedding and associated regularizers in facilitating SSGL.

Semi-supervised generative learning aims to learn the underlying class-conditional distribution of partially labeled data. Generative Adversarial Networks (GANs) have led to promising progress in this task. However, it still needs to further explore the issue of imbalance between real labeled data and fake data in the adversarial learning process. To address this issue, we propose a regularization technique based on Random Regional Replacement (R3-regularization) to facilitate the generative learning process. Specifically, we construct two types of between-class instances: cross-category ones and real-fake ones. These instances could be closer to the decision boundaries and are important for regularizing the classification and discriminative networks in our class-conditional GANs, which we refer to as R3-CGAN. Better guidance from these two networks makes the generative network produce instances with class-specific information and high fidelity. We experiment with multiple standard benchmarks, and demonstrate that the R3-regularization can lead to significant improvement in both classification and class-conditional image synthesis.

As a robust nonlinear similarity measure, the maximum correntropy criterion (MCC) has been successfully applied to active noise control (ANC) for impulsive noise. The default kernel function of the filtered-x maximum correntropy criterion (FxMCC) algorithm is the Gaussian kernel, which is desirable in many cases for its smooth and strict positive-definite. However, it is not always the best choice. In this study, a filtered-x generalized maximum correntropy criterion (FxGMCC) algorithm is proposed, which adopts the generalized Gaussian density (GGD) function as its kernel. The FxGMCC algorithm has greater robust ability against non-Gaussian environments, but, it still adopts a single error norm which exhibits poor convergence rate and noise reduction performance. To surmount this problem, an improved FxGMCC (IFxGMCC) algorithm with continuous mixed Lp-norm is proposed. Moreover, to make a trade-off between fast convergence rate and low steady-state misalignment, a convexly combined IFxGMCC (C-IFxGMCC) algorithm is further developed. The stability mechanism and computational complexity of the proposed algorithms are analyzed. Simulation results in the context of different impulsive noises as well as the real noise signals verify that the proposed algorithms are superior to most of the existing robust adaptive algorithms.

An individual-activation-factor memory proportionate affine projection algorithm (IAF-MPAPA) is proposed for sparse system identification in acoustic echo cancellation (AEC) scenarios. By utilizing an individual activation factor for each adaptive filter coefficient instead of a global activation factor, as in the standard proportionate affine projection algorithm (PAPA), the adaptation energy over the coefficients of the proposed IAF-MPAPA can achieve a better distribution, which leads to an improvement of the convergence performance. Moreover, benefiting from the memory characteristics of the proportionate coefficients, its computational complexity is less than the PAPA and improved PAPA (IPAPA). In the context of AEC and stereophonic AEC (SAEC) for highly sparse impulse responses, simulation results indicate that the proposed IAF-MPAPA outperforms the PAPA, IPAPA, and memory IPAPA (MIPAPA) in terms of the convergence rate and tracking capability when the unknown impulse response suddenly changes.

A novel complex-valued non-linear equaliser-based pipelined decision feedback recurrent neural network (CPDFRNN) is proposed in this study for non-linear channel equalisation in wireless communication systems. The CPDFRNN with low computational complexity, a modular structure comprising a number of modules that are interconnected in a chained form, is an extension of the recently proposed real-valued pipelined decision feedback recurrent neural equalisers. Each module is implemented by a small-scale complex-valued decision feedback recurrent neural network (CDFRNN). Moreover, a decision feedback part in each module can overcome the unstable characteristic of the complex-valued recurrent neural network (CRNN). To suit the modularity of the CPDFRNN, an adaptive amplitude complex-valued real-time recurrent learning (CRTRL) algorithm is presented. Simulations demonstrate that the CPDFRNN equaliser using the amplitude CRTRL algorithm with less computational complexity not only eliminates the adverse effects of the nesting architecture, but also provides a superior performance over the CRNN and CDFRNN equalisers for non-linear channels in wireless communication systems.

This correspondence presents an extended pipelined second-order Volterra (EPSOV) filter for active control of nonlinear noise processes. The corresponding nonlinear filtered-x algorithms using the filter bank implementation are also suggested. Compared to the standard SOV filter using the filtered-x least mean square (SOVFXLMS), those modules of the EPSOV filter can be performed simultaneously in a pipelined parallelism fashion, and this would lead to a significant improvement in its total computational efficiency. Results obtained from computer simulations for nonlinear noise processes demonstrate that the proposed method outperforms the SOV.

To combat the linear and non-linear distortions for time-invariant and time-variant channels, a novel adaptive joint process equaliser based on a pipelined decision feedback recurrent neural network (JPDFRNN) is proposed in this paper. The JPDFRNN consists of a number of simple small-scale decision feedback recurrent neural network (DFRNN) modules and a linear combiner. The cascaded DFRNN provides pre-processing for the linear combiner. Moreover, each DFRNN can provide a local interpolation for M sample points; the final linear combiner presents a global interpolation with good localisation properties. Furthermore, since those modules of non-linear subsection can be performed simultaneously in a pipelined parallelism fashion, this would result in a significant improvement in the total computational efficiency. Simulation results show that the performance of the JPDFRNN using the modified real-time recurrent learning (RTRL) algorithm is superior to that of the DFRNN and RNN for the non-linear time-invariant and time-variant channels.

To reduce the computational complexity of the bilinear recurrent neural network (BLRNN), a novel low-complexity nonlinear adaptive filter with a pipelined bilinear recurrent neural network (PBLRNN) is presented in this paper. The PBLRNN, inheriting the modular architectures of the pipelined RNN proposed by Haykin and Li, comprises a number of BLRNN modules that are cascaded in a chained form. Each module is implemented by a small-scale BLRNN with internal dynamics. Since those modules of the PBLRNN can be performed simultaneously in a pipelined parallelism fashion, it would result in a significant improvement of computational efficiency. Moreover, due to nesting module, the performance of the PBLRNN can be further improved. To suit for the modular architectures, a modified adaptive amplitude real-time recurrent learning algorithm is derived on the gradient descent approach. Extensive simulations are carried out to evaluate the performance of the PBLRNN on nonlinear system identification, nonlinear channel equalization, and chaotic time series prediction. Experimental results show that the PBLRNN provides considerably better performance compared to the single BLRNN and RNN models.

In this letter, a novel pipelined decision feedback RNN equalizer (PDFRNE) with low computational complexity is proposed. Since each module is a DFRNN with the decision feedback structure so that it can eliminate the past error remaining in the network. Moreover, the performance can be further improved. At the same time, it can overcome the unstableness due to its nature of the infinite impulse response (IIR) structure.