Robustness againstword substitution attacks iscrucial for text classifiers and fundamental to broader NLP robustness. Such attacks use semantically similar word substitutions. Existing certified defenses often compute loose outer bounds for the convex hull of word embeddings, including irrelevant words, degrading performance on clean and adversarial tests. Additionally, convex hull-based methods also struggle to emulate worst-case scenarios accurately. This paper proposes theMasked Adversarial Convex Combination(MACC) method, which models the solution space as a convex hull of word vectors and uses Variational Information Bottleneck theory to mask unnecessary words. We also introduce an empirical masking method based on the volume of the convex hull to enhance the performance. By reducing the number of words explored within the convex hull, MACC enables more precise mimicry of worst-case attacks. Experiments across models and datasets show MACC outperforms existing methods in clean and adversarial accuracy against word-level attacks.

Although deep learning has achieved great success in many computer vision tasks, its performance relies on the availability of large datasets with densely annotated samples. Such datasets are difficult and expensive to obtain. In this article, we focus on the problem of learning representation from unlabeled data for semantic segmentation. Inspired by two patch-based methods, we develop a novel self-supervised learning framework by formulating the jigsaw puzzle problem as a patch-wise classification problem and solving it with a fully convolutional network. By learning to solve a jigsaw puzzle comprising 25 patches and transferring the learned features to semantic segmentation task, we achieve a 5.8% point improvement on the Cityscapes dataset over the baseline model initialized from random values. It is noted that we use only about 1/6 training images of Cityscapes in our experiment, which is designed to imitate the real cases where fully annotated images are usually limited to a small number. We also show that our self-supervised learning method can be applied to different datasets and models. In particular, we achieved competitive performance with the state-of-the-art methods on the PASCAL VOC2012 dataset using significantly fewer time costs on pretraining.

Recent advances in low-power electronics, wireless communication, and bioengineering have produced many forms of implantable and migrating medical devices (a “migrating device” travels within the body along a specific route). Among them, smart implants and ingestible electronic pills that perform diagnostic functions inside the human body play increasing roles in modern medicine. Continuous access to physiological and biological information in real time disrupts the traditional symptom-based diagnostic approach, and ushers in a new era of technology- and evidence-based medicine. Despite the high potential impacts of diagnostic implants and migrating devices, numerous problems in both technical and biological domains must be solved. This talk focuses on several key problems common to these classes of devices: How to make them small while provide them with sufficient power and wireless communication functions when they are deeply located within the body? There are many compelling reasons to miniaturize these devices. First, smaller devices in centimeter or millimeter scales are easier to deliver in vivo using non-invasive (e.g., swallowing) or minimally invasive means, which in turn reduce medical care cost and complication rates. Second, smaller devices can be placed within less accessible compartments inside organs or tubular structures (e.g., a digestive tract) without interfering with the body’s normal biologic functions. Third, smaller devices improve biocompatibility and immune tolerance due to a smaller biological footprint. Despite these advantages, it remains technical challenge to construct such devices. For over two decades, our laboratory has been investigating the power and communication problems in miniature implants and migrating devices. While the power supply for a migrating device can be a battery if the lifetime of the device is short, a depleted battery within a permanent or long-term implant must be replaced surgically which is painful and expensive. To solve the power supply problem, we have studied three alternatives to the battery power. Firstly, the body’s own energy source is harvested using a biofuel cell taking advantage of the electrochemical activity of biological cells [1]. Secondly, electrical current is delivered to an implant through biological tissues using the volume conduction property of the human body [1]. Thirdly, wireless power transfer is used in which a power-carrying radio frequency wave (between megahertz and gigahertz) is transmitted from the outside to the inside of the human body [2]. To minimize device size and implement the data communication function, we have investigated a dual-functional design in which the wireless power delivery link and the data communication link are shared. Our design supports both downlink and uplink communication. While the downlink is relatively easy to implement using a power delivery signal with modulated amplitude, the uplink from the inside device to an outside unit is difficult, especially when the device is deeply located within the body. We must overcome the problems of limited power source, small transceiver antenna and long communication distance through layers of tissues. Our solution is to use a special tapped coil and a simple passive circuit to produce a strong pulsed magnetic field (PMF) for signal transmission within the body [2]. The strong PMF enables a data uplink from the implant to outside world in a high instantaneous transmission wattage through the shared power delivery channel, whereas a low average power consumption is maintained. Our experimental result showed that, when compared to the existing load shift keying (LSK) method, our method provides extraordinary improvements, with over 10 times increase in emitted signal strength and over 15 times increase in received signal strength.

Multilabel feature extraction (FE) is an effective preprocessing step to cope with some possible irrelevant, redundant, and noisy features, to reduce computational costs and even improve classification performance. Original normalized cross-covariance operator represents a kernel-based nonlinear dependence measure between features and labels, whose empirical estimator is formulated as a trace operation including two inverse matrices of feature and label kernels with a regularization constant. Due to such a complicated expression, it is impossible to derive an eigenvalue problem for linear FE directly. In this paper, we approximate this measure using Moore-Penrose inverse matrix, linear kernel for feature space, and delta kernel for label space, and then symmetrize the entire matrix in the trace operation, resulting in an effective approximated and symmetrized representation. According to orthonormal projection direction constraints, maximizing such a modified form induces a novel eigenvalue problem for multilabel linear FE. Experiments on 12 data sets illustrate that our proposed method works the best, compared with seven existing FE techniques, according to eight multilabel classification performance metrics and three statistical tests.

The rapid development of autonomous driving in recent years presents many challenges for scene understanding. As an essential step towards scene understanding, semantic segmentation has received increased attention in the past few years. Although deep learning based approaches have achieved great success in improving the segmentation accuracy, most of them suffer from an inefficiency problem and can hardly be applied to real-time applications. In this paper, we analyze the computational cost of Convolutional Neural Network (CNN) and find that the inefficiency of CNNs is mainly caused by their wide structure rather than deep structure. In addition, the success of pruning based model compression methods proves that there are many redundant channels in CNNs. Thus, we design a narrow while deep backbone network to improve the efficiency of semantic segmentation. By casting our network to fully convolutional network (FCN32) segmentation architecture, the basic structure of most segmentation methods, we achieve 61.5% mIoU on Cityscapes validation dataset with only 4.2G floating-point operations (FLOPs) on 1024×2048 inputs, which already outperforms one of the earliest real-time deep learning based segmentation methods: ENet (58.3% mIoU, 3.8G FLOPs on 640×360 inputs). By further refining the output resolution of our network to the 1/8 of the input resolution with a simple encoder-decoder structure, we achieve 65.3% mIoU on Cityscapes test set with 14.0G FLOPs and 39.9 frames per second (FPS) on Titan X card. We have made our model publicly available at https://github.com/zgyang-hnu/NDNet.

Social anxiety is a negative emotion which may impair the health of the heart and social functioning of an individual. This work analyzes the influence of social anxiety on the autonomic nerve control of the heart in two social exposure events: public speaking and thesis defending. In an experiment of public speaking, 59 human subjects were tested, and 11 conventional heartbeat measures and a heartbeat measure named the range of local Hurst exponents (RLHE) were evaluated for their capabilities to reveal the onset of social anxiety. Two-sample t-test between the baseline data and high anxiety data shows that social anxiety significantly reduces the complexity of the heartbeats. In an experiment of thesis defense, heartbeats data were acquired from nine graduate students. With the combination of three conventional features and the RLHE feature, a support vector machine classifier obtained true positive rate and true negative rate of 84.88 and 97.29 percent in the five-fold cross validation process of binary classification between high anxiety status and low anxiety status; the classifier also realized a generalization accuracy of 81.82 percent in detecting the high anxiety status in the thesis defense. A real-time anxiety monitoring system was established based on the above anxiety detecting method.

Semantic segmentation is a key step in scene understanding for autonomous driving. Although deep learning has significantly improved the segmentation accuracy, current highquality models such as PSPNet and DeepLabV3 are inefficient given their complex architectures and reliance on multi-scale inputs. Thus, it is difficult to apply them to real-time or practical applications. On the other hand, existing real-time methods cannot yet produce satisfactory results on small objects such as traffic lights, which are imperative to safe autonomous driving. In this paper, we improve the performance of real-time semantic segmentation from two perspectives, methodology and data. Specifically, we propose a real-time segmentation model coined Narrow Deep Network (NDNet) and build a synthetic dataset by inserting additional small objects into the training images. The proposed method achieves 65.7% mean intersection over union (mIoU) on the Cityscapes test set with only 8.4G floatingpoint operations (FLOPs) on 1024×2048 inputs. Furthermore, by re-training the existing PSPNet and DeepLabV3 models on our synthetic dataset, we obtained an average 2% mIoU improvement on small objects.

In the future, Human-Robot Interaction should be enabled by a compact, human-centered and ergonomic wearable device that can merge human and machine altogether seamlessly by constantly identifying each other's intentions. In this paper, we will showcase the use of an ergonomic and lightweight wearable device that can identify human's eye/facial gestures with physiological signal measurements. Since human's intentions are usually coupled with eye movements and facial expressions, through proper design of interactions using these gestures, we can let people interact with the robots or smart home objects naturally. Combined with Computer Vision object recognition algorithms, we can allow people use very simple and straightforward communication strategies to operate telepresence robot and control smart home objects remotely, totally “Hands-Free”. People can wear a VR head-mounted display and see through the robot's eyes (the remote camera attached on the robot) and interact with the smart home devices intuitively by simple facial gestures or blink of the eyes. It is tremendous beneficial for the people with motor impairment as an assistive tool. For the normal people without disabilities, they can also free their hands to do other tasks and operate the smart home devices at the same time as multimodal control strategies.

Energy harvesters are becoming increasingly popular as power sources for IoT edge devices. However, one of the intrinsic problems of energy harvester is that harvesting power is often weak and frequently interrupted. Therefore, energy harvesting powered edge devices have to work intermittently. To maintain execution progress, execution states need to be checkpointed into the non-volatile memory before each power failure. In this way, previous execution states can be resumed after power comes back again. Nevertheless, frequent checkpointing and low charging efficiency generate significant energy overhead. To alleviate these problems, this article conducts a thorough energy efficiency analysis and proposes three algorithms to maximize the energy efficiency of program execution. First, a non-volatile processor-aware task scheduling algorithm is proposed to reduce the size of checkpointing data. Second, a tentative checkpointing avoidance technique is proposed to avoid checkpointing for further reduction of checkpointing overhead. Finally, a dynamic wake-up strategy is proposed to wake up the edge device at proper voltages where the total hardware and software overhead is minimized for further energy efficiency maximization. The experiments on a real testbed demonstrate that, with the proposed algorithms, an edge device is resilient to the extremely weak and intermittent power supply and the energy efficiency can be achieved more than 2× higher than the fundamental baseline and 1.5× higher than the state-of-the-art technique.

Efficient implementation of finite field multipliers based on a reordered normal basis (RNB) is highly desirable in the current/emerging cryptosystems since it offers almost free realization of squaring operation. Therefore, in this paper, we propose novel bit-parallel and digit-serial finite field multipliers over GF(2m) based on RNB. By efficient transformation of the core multiplication algorithm using a unique circular shifting feature, we have derived an efficient algorithm for low-complexity systolic mapping. Both bit-parallel and digit-serial structures of the multipliers are then obtained and optimized to enhance the area-time efficiency. We have also utilized the unique feature of the proposed multiplication algorithm to obtain the systolic multipliers by Karatsubalike decomposition. Detailed analysis and comparison show the superior performance of the proposed implementation. For example, the proposed regular and Karatsuba-based bit-parallel designs involve at least 48.4% less area-delay product (ADP) and 42.2% less power-delay product (PDP) than the best existing ones (37.7% and 55.3% less ADP and PDP on field-programmable gate array platform), respectively. The proposed multipliers, because of their lower area-time complexities, can be used for efficient realization of cryptographic applications.

In the era of network service, the user authentication becomes more indispensable but also vulnerable. Traditional user verification approaches such as PIN or pattern lock suffer from easy hacking and replica, motivating the research on many new approaches like gesture-based security. Compare to traditional authentications, the gesture-based security utilizes the user interacts with the device as a dynamic authentication pattern in real-time, offering higher complexity and better reliability. However, gesture-based security still lacks sufficient research on data sampling and preprocessing techniques on classification accuracy. In this work, we develop SlowMo, a novel gesture security technique for user classification in low sampling-rate environments. The proposed algorithm provides maximum classification accuracy at a sampling rate of 4Hz with extreme low power consumption suggesting a more capable adaptation to the security environment. It can achieve classification accuracy as high as 89% with power consumption negligible to the user.

Films such as Robocop, The Matrix, and Pacific Rim have explored the possibilities of using Brain-Computer Interfaces (BCIs) to control machines with only thought. In this paper, we enhance the power of thought through a novel method of combining the powers of the brain wave, the most influential signal in the human body, and the wide range of information found on the internet during today's digital age. Because brain wave signals are affected by human thought, certain patterns of EEG (Electroencephalogram) signals can be associated with specific actions or ideas. By analyzing these characteristics of brain wave signals through powerful EEG capturing devices, we were able to create a user interface where a human can have access to a variety of internet services, including search results, social media and IFTTT, by merely thinking of or focusing on a word or phrase on a computer screen.

Recent advances in mobile devices have made profound changes in people's daily lives. In particular, the impact of easy access of information by the smartphone has been tremendous. However, the impact of mobile devices on healthcare has been limited. Diagnosis and treatment of diseases are still initiated by occurrences of symptoms, and technologies and devices that emphasize on disease prevention and early detection outside hospitals are under-developed. Besides healthcare, mobile devices have not yet been designed to fully benefit people with special needs, such as the elderly and those suffering from certain disabilities, such blindness. In this paper, an overview of our research on a new wearable computer called eButton is presented. The concepts of its design and electronic implementation are described. Several applications of the eButton are described, including evaluating diet and physical activity, studying sedentary behavior, assisting the blind and visually impaired people, and monitoring older adults suffering from dementia.

This paper proposes a compact configuration of aircraft flows at intersections. The goal is to achieve a higher capacity of the airspace, allowing more aircraft to safely fly through a fixed region. Intersections of aircraft flows can be considered as basic building blocks for air traffic networks, and traffic networks can be designed through finding optimal arrangements of intersections whose conflict zones do not overlap. A conflict zone is defined as a minimal circular area centered at the intersection of two flows, which allows aircraft approaching the intersections to resolve conflict completely within the conflict zone. This paper derives the relationship between the size of a conflict zone and the intersection angle of the two flows. Such a relationship guides the choice of the most compact configuration for intersecting aircraft flows. An example involving multiple converging flows of aircraft demonstrates the efficiency of the proposed configuration of intersections. The result of conflict resolution shows a greatly reduced traffic complexity. Therefore, our study provides a potential solution to increase airspace capacity.

This paper proposes a method based on the Artificial societies, computational experiments, and Parallel execution (ACP) approach to build parallel public transportation systems (PPTSs). The framework and components of a PPTS are analyzed, and some details for building the PPTS are discussed. One prototype based on intelligent traffic clouds is established. One specific PPTS is developed for the Guangzhou 2010 Asian Games in the case study, and its effectiveness is verified through the evaluation of two evacuation plans for the Asian Games.

Intelligent mobile devices have been widely serving in almost all aspects of everyday life, spanning from communication, web surfing, entertainment, to daily organizer. A large amount of sensitive and private information is stored on the mobile device, leading to severe data security concern. In this work, we propose a novel mobile user classification and authorization scheme based on the recognition of user's gesture. Compared to other security solutions like password, track pattern and finger print etc., our scheme can continuously evolve for better protection during the usage cycle of the mobile device. Besides the regular interactive screen and sensors of modern mobile devices, our scheme does not require any additional hardware supports.

The invention of neuromorphic computing architecture is inspired by the working mechanism of human-brain. Memristor technology revitalized neuromorphic computing system design by efficiently executing the analog Matrix-Vector multiplication on the memristor-based crossbar (MBC) structure. However, programming the MBC to the target state can be very challenging due to the difficulty to real-time monitor the memristor state during the training. In this work, we quantitatively analyzed the sensitivity of the MBC programming to the process variations and input signal noise. We then proposed a noise-eliminating training method on top of a new crossbar structure to minimize the noise accumulation during the MBC training and improve the trained system performance, i.e., the pattern recall rate. A digital-assisted initialization step for MBC training is also introduced to reduce the training failure rate as well as the training time. Experimental results show that our noise-eliminating training method can improve the pattern recall rate. For the tested patterns with 128 x 128 pixels our technique can reduce the MBC training time by 12.6% ~ 14.1% for the same pattern recognition rate, or improve the pattern recall rate by 18.7% ~ 36.2% for the same training time.

The complexity of human-machine interaction (HMI) is rapidly growing in modern industrial, medical, and military systems. Human operators are often challenged by control of high-order systems or unstable systems near the limits of controllability. However, there is no quantitative indication of human performance and cognitive workload in these difficult HMI tasks. Here, we characterize HMI as information flows measured in bits per second (b/s). We propose a control-theoretic framework to estimate the rate of information transmission in manual control. We demonstrate our method in an example of stabilizing an inverted pendulum, where we derive that, for a normal human operator, the information-transmission rate of manual control with one degree of freedom ranges between 3 and 4 b/s. This quantitative indication reveals the potential and limitation of human manual control and is instructive to the design of HMI interfaces that may maximally utilize human control commands.

In this study, joint source–channel decoding for non-binary source samples is conducted. The non-binary source samples can be modelled as the output of a multi-state Markov chain (MC). As the source samples are directly transmitted after channel coding without source compression, the transmitted signals can be highly correlated. At the receiver, the multi-state MC module can be designed to exploit the statistical correlation of source samples to improve the error correcting performance. However, as the number of states is increased, the multi-state MC module requires high computational complexity. To alleviate this problem, a simplified MC module is proposed. In the simplified MC module, the multi-state MC is replaced with multiple number of two-state MCs each of which exploits bit-level correlation of samples. Simulation results demonstrate that the simplified MC module can lead to competitive reduction in the required signal-to-noise ratio in comparison with the multi-state MC module with reduced computational complexity.

This paper demonstrates a synergy-based brain-machine interface that uses low-dimensional command signals to control a high dimensional virtual hand. First, temporal postural synergies were extracted from the angular velocities of finger joints of five healthy subjects when they performed hand movements that were similar to activities of daily living. Two synergies inspired from the extracted synergies, namely, two-finger pinch and whole-hand grasp, were used in real-time brain control, where a virtual hand with 10 degrees of freedom was controlled to grasp or pinch virtual objects. These two synergies were controlled by electrocorticographic (ECoG) signals recorded from two electrodes of an electrode array that spanned motor and speech areas of an individual with intractable epilepsy, thus demonstrating closed loop control of a synergy-based brain-machine interface.

Human factors constitute a core aspect of air traffic control concepts and must be addressed when dealing with traffic flow stability problems. This paper aims to evaluate how the usage of airspace depends on the degree of flexibility in pilot decision-making. Specifically, we study the airspace usage of two intersecting flows of aircraft when pilot maneuver preferences vary. The amount of airspace required for the two flows to cross without conflict can reveal certain aspects of traffic complexity of this specific traffic pattern. To simplify the analysis, we use models of instantaneous lateral and longitudinal position changes to approximate the heading-change and speed-change maneuvers of the aircraft, respectively. Pilot preferences, which are expressed as preferred heading or speed changes, are reflected by different penalty functions of position displacements that the pilots attempt to minimize during conflict resolution. Under the same pilot preferences, the aircraft flow stability is preserved using a decentralized conflict-resolution scheme. However, when the pilot preferences are allowed to vary so that individual aircraft have more control freedom in conflict resolution, the intersecting flows require a much larger fraction of airspace to ensure conflict-free and stable flows of aircraft.

Postural synergies of the hand have been widely proposed in the literature, but only a few attempts were made to visualize temporal postural synergies, i.e., profiles of postural synergies varying over time. This paper aims to derive temporal postural synergies from kinematic synergies extracted from joint angular velocity profiles of rapid grasping movements. The rapid movements constrain the kinematic synergies to combine instantaneously, and thus, the movements can be approximated by a weighted summation of synchronous synergies. After being extracted by using singular value decomposition, the synchronous kinematic synergies were translated into temporal postural synergies, which revealed strategies of enslaving, metacarpal flexion for larger movements, and hierarchical recruitment of joints, adapted by subjects while grasping.

This paper provides an artifact-pattern-matching framework and mathematical model to analyze the dynamic behaviors of the SOA solution design in model driven fashion and provide recommendations for optimal solution pattern enablement for solution artifacts. The artifact-pattern-matching system can be dynamically tuned based on the practitionerspsila final selections of there commendations. Specifically, we propose a set of solution patterns to guide SOA solution architects through the process of consuming and configuring SOA artifacts for composing SOA solutions. The resulting multi-dimensional cascading flagging method is also presented in this paper. As an example, impact analysis patterns are used as solution patterns to support traceability enablement. We present some future directions of leveraging reinforcement learning algorithms to enrich the design quality analytics of SOA solution.

This paper proposes an efficient method to acquire sources of tremor in patients with movement disorders based on blind source separation of convolutive mixtures. The extracted sources indicated neural activities that might be generated in the central nervous system. Four patients with essential tremor were tested in a set of movement tasks. Subjects wore a data glove that measured finger movements of the hand. The experimental data were then fed to a convolutive-mixture model, which revealed sources that imbibed in them the tremor frequency components of 2--8 Hz. Time--frequency analysis of these sources might be of potential help to clinicians to devise tasks that can manifest visible tremor from patients.

This paper aims to provide reliable indications of driver drowsiness based on the characteristics of driver-vehicle interaction. A test bed was built under a simulated driving environment, and a total of 12 subjects participated in two experiment sessions requiring different levels of sleep (partial sleep-deprivation versus no sleep-deprivation) before the experiment. The performance of the subjects was analyzed in a series of stimulus-response and routine driving tasks, which revealed the performance differences of drivers under different sleep-deprivation levels. The experiments further demonstrated that sleep deprivation had greater effect on rule-based than on skill-based cognitive functions: when drivers were sleep-deprived, their performance of responding to unexpected disturbances degraded, while they were robust enough to continue the routine driving tasks such as lane tracking, vehicle following, and lane changing. In addition, we presented both qualitative and quantitative guidelines for designing drowsy-driver detection systems in a probabilistic framework based on the paradigm of Bayesian networks. Temporal aspects of drowsiness and individual differences of subjects were addressed in the framework.

This paper studies the conflict resolution for intersecting flows of mobile agents based on planar space partition. The idea of space partition is first demonstrated for two intersecting flows of mobile agents. Then, for three intersecting flows, where simple decentralized conflict avoidance rules may not handle all traffic scenarios, it is proved that certain periodic partitions of space are able to provide conflict resolution for any distribution of agents in the flows. A computational procedure based on mixed integer programming is further proposed to find optimal space partitions. The approach of space partition is not an online optimization algorithm. An online algorithm may find optimal resolution of conflict for a specific set of mobile agents but has to be rerun each time when new agents arrive, whereas a periodic partition of space provides a priori geometrical configuration for conflict avoidance regardless of the number and arriving patterns of the agents. Moreover, the offline nature of space partition does not imply a decrease of performance. As demonstrated in an example involving three symmetrically arranged agent flows, the optimal space partition has found a tight upper bound for the magnitude of any conflict-free maneuvers.

This paper studies the behavior of recurrent neural networks with lateral inhibition. Such network architecture is important in biological neural systems. General conditions determining the existence, number, and stability of network equilibria are derived. The manner in which these features depend upon steepness of neuronal activation functions and the strength of lateral inhibition is demonstrated for a broad range of nondecreasing activation functions including the discontinuous threshold function which represents the infinite gain limit. For uniform lateral inhibitory networks, the lateral inhibition is shown to sharpen neuron output patterns by increasing separation of suprathreshold activity levels of competing neurons. This results in the tendency of one neuron's output to dominate those of the others which can afford a "winner-take-all" (WTA) mechanism. Importantly, multiple stable equilibria may exist and shifts in inputs levels may yield network state transitions that exhibit hysteresis. A limitation of using lateral inhibition to implement WTA is further demonstrated. The possible significance of these identified network dynamics to physiology and pathophysiology of the striatum (particularly in Parkinsonian rest tremor) is discussed.

This paper analyzes two intersecting flows of aircraft that must avoid each other, utilizing maneuvers modeled by instantaneous heading changes. Sufficient conditions are derived for stable conflict avoidance, whereby one aircraft's conflict-avoidance action does not generate cascaded diverging conflict-avoidance behavior (known as the domino effect) in neighboring aircraft. The mathematics involved with the heading-change model are considerably more complex than those using the simpler offset-maneuver models utilized in the previous work. However, this additional effort is largely compensated by the improved realism of the maneuver model, as well as new insights in the inherent robustness of conflict-avoidance maneuver schemes based on heading changes against specified separation standards. The analytical results are compared with simulations.

This letter studies the properties of the random neural networks (RNNs) with state-dependent firing neurons. It is assumed that the times between successive signal emissions of a neuron are dependent on the neuron potential. Under certain conditions, the networks keep the simple product form of stationary solutions and exhibit enhanced capacity of adjusting the probability distribution of the neuron states. It is demonstrated that desired associative memory states can be stored in the networks.

This paper considers the problem of two intersecting aircraft flows under decentralized conflict resolution rules. Considering aircraft flowing through a fixed control volume, new air traffic control models and scenarios are defined that enable the study of long-term aircraft flow stability. For a class of two intersecting aircraft flows, this paper considers conflict scenarios involving arbitrary encounter angles. It is shown that aircraft flow stability, defined both in terms of safety and performance, is preserved under the decentralized conflict resolution algorithm considered. It is shown that the lateral deviations experienced by aircraft in each flow are bounded.

This paper examines the function approximation properties of the "random neural-network model" or GNN. The output of the GNN can be computed from the firing probabilities of selected neurons. We consider a feedforward Bipolar GNN (BGNN) model which has both "positive and negative neurons" in the output layer, and prove that the BGNN is a universal function approximator. Specifically, for any f is an element of C([0, 1]s) and any epsilon>0, we show that there exists a feedforward BGNN which approximates f uniformly with error less than epsilon. We also show that after some appropriate clamping operation on its output, the feedforward GNN is also a universal function approximator.

This paper reports on a related study on approximation theory of fuzzy systems. First, some basic principles are presented to construct membership functions. Then, an approach is proposed to form membership functions by using translations and dilations of one fixed function (called a basis function) which is very similar to that in wavelets analysis. The properties of this type of membership function reflect the advantages of the given approach. Finally, it is proved that fuzzy systems based on such membership functions are universal approximators under certain mild conditions on the basis function. This conclusion expands the family of fuzzy systems which can be universal approximators.