Self-Organizing Type-2 Fuzzy Double Loop Recurrent Neural Network for Uncertain Nonlinear System Control.

Nonlinear systems, such as robotic systems, play an increasingly important role in our modern daily life and have become more dominant in many industries; however, robotic control still faces various challenges due to diverse and unstructured work environments. This article proposes a double-loop recurrent neural network (DLRNN) with the support of a Type-2 fuzzy system and a self-organizing mechanism for improved performance in nonlinear dynamic robot control. The proposed network has a double-loop recurrent structure, which enables better dynamic mapping. In addition, the network combines a Type-2 fuzzy system with a double-loop recurrent structure to improve the ability to deal with uncertain environments. To achieve an efficient system response, a self-organizing mechanism is proposed to adaptively adjust the number of layers in a DLRNN. This work integrates the proposed network into a conventional sliding mode control (SMC) system to theoretically and empirically prove its stability. The proposed system is applied to a three-joint robot manipulator, leading to a comparative study that considers several existing control approaches. The experimental results confirm the superiority of the proposed system and its effectiveness and robustness in response to various external system disturbances.

Decoder Choice Network for Metalearning.

Metalearning has been widely applied for implementing few-shot learning and fast model adaptation. Particularly, existing metalearning methods have been exploited to learn the control mechanism for gradient descent processes, in an effort to facilitate gradient-based learning in gaining high speed and generalization ability. This article presents a novel method that controls the gradient descent process of the model parameters in a neural network, by limiting the model parameters within a low-dimensional latent space. The main challenge for implementing this idea is that a decoder with many parameters may be required. To tackle this problem, the article provides an alternative design of the decoder with a structure that shares certain weights, thereby reducing the number of required parameters. In addition, this work combines ensemble learning with the proposed approach to improve the overall learning performance. Systematic experimental studies demonstrate that the proposed approach offers results superior to the state of the art in performing the Omniglot classification and miniImageNet classification tasks.

Solving Robotic Trajectory Sequential Writing Problem via Learning Character's Structural and Sequential Information.

The writing sequence of numerals or letters often affects aesthetic aspects of the writing outcomes. As such, it remains a challenge for robotic calligraphy systems to perform, mimicking human writers' implicit intention. This article presents a new robot calligraphy system that is able to learn writing sequences with limited sequential information, producing writing results compatible to human writers with good diversity. In particular, the system innovatively applies a gated recurrent unit (GRU) network to generate robotic writing actions with the support of a prelabeled trajectory sequence vector. Also, a new evaluation method is proposed that considers the shape, trajectory sequence, and structural information of the writing outcome, thereby helping ensure the writing quality. A swarm optimization algorithm is exploited to create an optimal set of parameters of the proposed system. The proposed approach is evaluated using Arabic numerals, and the experimental results demonstrate the competitive writing performance of the system against state-of-the-art approaches regarding multiple criteria (including FID, MAE, PSNR, SSIM, and PerLoss), as well as diversity performance concerning variance and entropy. Importantly, the proposed GRU-based robotic motion planning system, supported with swarm optimization can learn from a small dataset, while producing calligraphy writing with diverse and aesthetically pleasing outcomes.

Modality independent adversarial network for generalized zero shot image classification.

Zero Shot Learning (ZSL) aims to classify images of unseen target classes by transferring knowledge from source classes through semantic embeddings. The core of ZSL research is to embed both visual representation of object instance and semantic description of object class into a joint latent space and learn cross-modal (visual and semantic) latent representations. However, the learned representations by existing efforts often fail to fully capture the underlying cross-modal semantic consistency, and some of the representations are very similar and less discriminative. To circumvent these issues, in this paper, we propose a novel deep framework, called Modality Independent Adversarial Network (MIANet) for Generalized Zero Shot Learning (GZSL), which is an end-to-end deep architecture with three submodules. First, both visual feature and semantic description are embedded into a latent hyper-spherical space, where two orthogonal constraints are employed to ensure the learned latent representations discriminative. Second, a modality adversarial submodule is employed to make the latent representations independent of modalities to make the shared representations grab more cross-modal high-level semantic information during training. Third, a cross reconstruction submodule is proposed to reconstruct latent representations into the counterparts instead of themselves to make them capture more modality irrelevant information. Comprehensive experiments on five widely used benchmark datasets are conducted on both GZSL and standard ZSL settings, and the results show the effectiveness of our proposed method.

Type-2 Fuzzy Hybrid Controller Network for Robotic Systems.

Dynamic control, including robotic control, faces both the theoretical challenge of obtaining accurate system models and the practical difficulty of defining uncertain system bounds. To facilitate such challenges, this paper proposes a control system consisting of a novel type of fuzzy neural network and a robust compensator controller. The new fuzzy neural network is implemented by integrating a number of key components embedded in a Type-2 fuzzy cerebellar model articulation controller (CMAC) and a brain emotional learning controller (BELC) network, thereby mimicking an ideal sliding mode controller. The system inputs are fed into the neural network through a Type-2 fuzzy inference system (T2FIS), with the results subsequently piped into sensory and emotional channels which jointly produce the final outputs of the network. That is, the proposed network estimates the nonlinear equations representing the ideal sliding mode controllers using a powerful compensator controller with the support of T2FIS and BELC, guaranteeing robust tracking of the dynamics of the controlled systems. The adaptive dynamic tuning laws of the network are developed by exploiting the popular brain emotional learning rule and the Lyapunov function. The proposed system was applied to a robot manipulator and a mobile robot, demonstrating its efficacy and potential; and a comparative study with alternatives indicates a significant improvement by the proposed system in performing the intelligent dynamic control.

Interaction-Based Human Activity Comparison.

Traditional methods for motion comparison consider features from individual characters. However, the semantic meaning of many human activities is usually defined by the interaction between them, such as a high-five interaction of two characters. There is little success in adapting interaction-based features in activity comparison, as they either do not have a fixed topology or are in high dimensional. In this paper, we propose a unified framework for activity comparison from the interaction point of view. Our new metric evaluates the similarity of interaction by adapting the Earth Mover's Distance onto a customized geometric mesh structure that represents spatial-temporal interactions. This allows us to compare different classes of interactions and discover their intrinsic semantic similarity. We created five interaction databases of different natures, covering both two-characters (synthetic and real-people) and character-object interactions, which are open for public uses. We demonstrate how the proposed metric aligns well with the semantic meaning of the interaction. We also apply the metric in interaction retrieval and show how it outperforms existing ones. The proposed method can be used for unsupervised activity detection in monitoring systems and activity retrieval in smart animation systems.

Multi-criterion mammographic risk analysis supported with multi-label fuzzy-rough feature selection.

CONTEXT AND BACKGROUND: Breast cancer is one of the most common diseases threatening the human lives globally, requiring effective and early risk analysis for which learning classifiers supported with automated feature selection offer a potential robust solution. MOTIVATION: Computer aided risk analysis of breast cancer typically works with a set of extracted mammographic features which may contain significant redundancy and noise, thereby requiring technical developments to improve runtime performance in both computational efficiency and classification accuracy. HYPOTHESIS: Use of advanced feature selection methods based on multiple diagnosis criteria may lead to improved results for mammographic risk analysis. METHODS: An approach for multi-criterion based mammographic risk analysis is proposed, by adapting the recently developed multi-label fuzzy-rough feature selection mechanism. RESULTS: A system for multi-criterion mammographic risk analysis is implemented with the aid of multi-label fuzzy-rough feature selection and its performance is positively verified experimentally, in comparison with representative popular mechanisms. CONCLUSIONS: The novel approach for mammographic risk analysis based on multiple criteria helps improve classification accuracy using selected informative features, without suffering from the redundancy caused by such complex criteria, with the implemented system demonstrating practical efficacy.

A Deep-Learning-Driven Light-Weight Phishing Detection Sensor.

This paper designs an accurate and low-cost phishing detection sensor by exploring deep learning techniques. Phishing is a very common social engineering technique. The attackers try to deceive online users by mimicking a uniform resource locator (URL) and a webpage. Traditionally, phishing detection is largely based on manual reports from users. Machine learning techniques have recently been introduced for phishing detection. With the recent rapid development of deep learning techniques, many deep-learning-based recognition methods have also been explored to improve classification performance. This paper proposes a light-weight deep learning algorithm to detect the malicious URLs and enable a real-time and energy-saving phishing detection sensor. Experimental tests and comparisons have been conducted to verify the efficacy of the proposed method. According to the experiments, the true detection rate has been improved. This paper has also verified that the proposed method can run in an energy-saving embedded single board computer in real-time.

An Improved Fuzzy Brain Emotional Learning Model Network Controller for Humanoid Robots.

The brain emotional learning (BEL) system was inspired by the biological amygdala-orbitofrontal model to mimic the high speed of the emotional learning mechanism in the mammalian brain, which has been successfully applied in many real-world applications. Despite of its success, such system often suffers from slow convergence for online humanoid robotic control. This paper presents an improved fuzzy BEL model (iFBEL) neural network by integrating a fuzzy neural network (FNN) to a conventional BEL, in an effort to better support humanoid robots. In particular, the system inputs are passed into a sensory and emotional channels that jointly produce the final outputs of the network. The non-linear approximation ability of the iFBEL is achieved by taking the BEL network as the emotional channel. The proposed iFBEL works with a robust controller in generating the hand and gait motion of a humanoid robot. The updating rules of the iFBEL-based controller are composed of two parts, including a sensory channel followed by the updating rules of the conventional BEL model, and the updating rules of the FNN and the robust controller which are derived from the "Lyapunov" function. The experiments on a three-joint robot manipulator and a six-joint biped robot demonstrated the superiority of the proposed system in reference to a conventional proportional-integral-derivative controller and a fuzzy cerebellar model articulation controller, based on the more accurate and faster control performance of the proposed iFBEL.

A Developmental Learning Approach of Mobile Manipulator via Playing.

Inspired by infant development theories, a robotic developmental model combined with game elements is proposed in this paper. This model does not require the definition of specific developmental goals for the robot, but the developmental goals are implied in the goals of a series of game tasks. The games are characterized into a sequence of game modes based on the complexity of the game tasks from simple to complex, and the task complexity is determined by the applications of developmental constraints. Given a current mode, the robot switches to play in a more complicated game mode when it cannot find any new salient stimuli in the current mode. By doing so, the robot gradually achieves it developmental goals by playing different modes of games. In the experiment, the game was instantiated into a mobile robot with the playing task of picking up toys, and the game is designed with a simple game mode and a complex game mode. A developmental algorithm, "Lift-Constraint, Act and Saturate," is employed to drive the mobile robot move from the simple mode to the complex one. The experimental results show that the mobile manipulator is able to successfully learn the mobile grasping ability after playing simple and complex games, which is promising in developing robotic abilities to solve complex tasks using games.

Fano-resonance-based ultra-high-resolution ratio-metric wavelength monitor on silicon.

An integrated ultra-high-resolution ratio-metric wavelength monitor (RMWM) with compact size based on slope tunable Fano resonance is demonstrated on silicon. The Fano resonance is generated by adding an asymmetric microring inside and coupling with the outer ring to produce a nonlinear phase shift. The slope tunability is achieved by controlling the microheaters to adjust the phase condition. Two asymmetric embedded microring resonators (AEMR) are functioned as edge filters and designed to achieve an "X-type" spectral response in a particular wavelength range. An ultra-high resolution of 0.8 pm in a 0.47 nm wide wavelength range is experimentally demonstrated. This device could be applied in on-chip high-sensitivity wavelength monitoring sensing.

Temperature-fluctuation-sensitive accumulative effect of the phase measurement errors in low-coherence interferometry in characterizing arrayed waveguide gratings.

We investigate the accumulative effect of the phase measurement errors in characterizing optical multipath components by low-coherence interferometry. The accumulative effect is caused by the fluctuation of the environment temperature, which leads to the variation of the refractive index of the device under test. The resulting phase measurement errors accumulate with the increasing of the phase difference between the two interferometer arms. Our experiments were carried out to demonstrate that the accumulative effect is still obvious even though the thermo-optical coefficient of the device under test is quite small. Shortening the measurement time to reduce the fluctuation of the environment temperature can effectively restrain the accumulative effect. The experiments show that when the scanning speed increases to 4.8 mm/s, the slope of the phase measurement errors decreases to 5.52×10(-8), which means the accumulative effect can be ignored.

An all-optical modulation method in sub-micron scale.

We report a theoretical study showing that by utilizing the illumination of an external laser, the Surface Plasmon Polaritons (SPP) signals on the graphene sheet can be modulated in the sub-micron scale. The SPP wave can propagate along the graphene in the middle infrared range when the graphene is properly doped. Graphene's carrier density can be modified by a visible laser when the graphene sheet is exfoliated on the hydrophilic SiO2/Si substrate, which yields an all-optical way to control the graphene's doping level. Consequently, the external laser beam can control the propagation of the graphene SPP between the ON and OFF status. This all-optical modulation effect is still obvious when the spot size of the external laser is reduced to 400 nm while the modulation depth is as high as 114.7 dB/µm.

Microring-based ratio-metric wavelength monitor on silicon.

An integrated dual-ring ratio-metric wavelength monitor (DR-RMWM) with ultra-high wavelength resolution and compact size is demonstrated. Two microrings are used as the edge filters and designed to achieve an "X-type" spectral response in a particular wavelength range. It is fabricated with the CMOS-compatible fabrication process on the silicon-on-insulator. The measured wavelength resolution is 5 pm in a 1.2 nm wide wavelength range. By tuning the resonant wavelength thermally, the functional wavelength range can be shifted. The DR-RMWM can find applications in wavelength monitoring systems, especially the on-chip systems.

Ultracompact optical modulator based on graphene-silica metamaterial.

We present an ultracompact electro-optic modulator with a length of 15 nm by utilizing a permittivity-tunable metamaterial channel, which is composed of alternative layers of graphene and silica. Optical waves can propagate through the metamaterial channel only if its permittivity is tuned to be near zero. The finite-difference time-domain (FDTD) simulations show the insertion loss is roughly -0.27 dB while the 3 dB extinction ratio is obtained with a 0.8 V gate voltage. The device's footprint is as small as 0.01 µm2. This modulator consumes low power and can potentially be ultrafast.

Low-chirp high-extinction-ratio modulator based on graphene-silicon waveguide.

We present a hybrid graphene-silicon waveguide, which consists of a lateral slot waveguide with three layers of graphene flakes inside. Through a theoretical analysis, an effective index variation for about 0.05 is found in the waveguide by applying a voltage on the graphene. We designed a Mach-Zehnder modulator based on this waveguide and demonstrated it can process signals nearly chirp-free. The calculation shows that the driving voltage is only 1 V even if the length of the arm is shortened to be 43.54 µm. An extinction up to 34.7 dB and a minimum chirp parameter of -0.006 are obtained. Its insertion loss is roughly -1.37 dB. This modulator consumes low power and has a small footprint. It can potentially be ultrafast as well as CMOS compatible.

Towards a Fuzzy Expert System on Toxicological Data Quality Assessment.

Quality assessment (QA) requires high levels of domain-specific experience and knowledge. QA tasks for toxicological data are usually performed by human experts manually, although a number of quality evaluation schemes have been proposed in the literature. For instance, the most widely utilised Klimisch scheme1 defines four data quality categories in order to tag data instances with respect to their qualities; ToxRTool2 is an extension of the Klimisch approach aiming to increase the transparency and harmonisation of the approach. Note that the processes of QA in many other areas have been automatised by employing expert systems. Briefly, an expert system is a computer program that uses a knowledge base built upon human expertise, and an inference engine that mimics the reasoning processes of human experts to infer new statements from incoming data. In particular, expert systems have been extended to deal with the uncertainty of information by representing uncertain information (such as linguistic terms) as fuzzy sets under the framework of fuzzy set theory and performing inferences upon fuzzy sets according to fuzzy arithmetic. This paper presents an experimental fuzzy expert system for toxicological data QA which is developed on the basis of the Klimisch approach and the ToxRTool in an effort to illustrate the power of expert systems to toxicologists, and to examine if fuzzy expert systems are a viable solution for QA of toxicological data. Such direction still faces great difficulties due to the well-known common challenge of toxicological data QA that "five toxicologists may have six opinions". In the meantime, this challenge may offer an opportunity for expert systems because the construction and refinement of the knowledge base could be a converging process of different opinions which is of significant importance for regulatory policy making under the regulation of REACH, though a consensus may never be reached. Also, in order to facilitate the implementation of Weight of Evidence approaches and in silico modelling proposed by REACH, there is a higher appeal of numerical quality values than nominal (categorical) ones, where the proposed fuzzy expert system could help. Most importantly, the deriving processes of quality values generated in this way are fully transparent, and thus comprehensible, for final users, which is another vital point for policy making specified in REACH. Case studies have been conducted and this report not only shows the promise of the approach, but also demonstrates the difficulties of the approach and thus indicates areas for future development.

Characteristics of electro-refractive modulating based on Graphene-Oxide-Silicon waveguide.

Graphene has attracted a high level of research interest because of its outstanding electronic transport properties and optical properties. Based on the Kubo formalism and the Maxwell equations, it's demonstrated that the optical conductivity of graphene can be controlled through the applied voltage. And we find that the graphene-oxide-silicon (GOS) based waveguide can be made into either the electro-absorptive or electron-refractive modulators. Using graphene as the active medium, we present a new electro-refractive Mach-Zender interferometer based on the GOS structure. This new GOS-based electron-refractive modulation mechanism can enable novel architectures for on-chip optical communications.