Learning-based Mechanisms Research Articles

Traffic Prediction for Intelligent Transportation System using Deep Learning

The most important challenge to sustainable mobility is persistent congestions of differing strength and duration in the dense transport networks. The standard Adaptive Traffic Signal Control cannot properly address this kind of congestion. Deep learning-based mechanisms have proved their significance to anticipate in adjective outcomes to improve the decision making on the predictions of traffic length. The deep learning models have long been used in many application domains which needed the identification and prioritization of adverse factors for a simplifying human life. Several methods are being popularly used to handle real time problems occurring from traffic congestion. This study demonstrates the capability of DL models to overcome the traffic congestion by simply allowing the vehicles through a signal depending on the length of vehicles. Our proposed method integrates a numeral of approach, intended to advance the cooperativeness of the explore operation. In this work, we implement the application to detect the number of vehicles in the images from the user and gives vehicles counts. To detect the vehicles count here we are using the YOLO pretrained weights.

A Levenberg–Marquardt Based Neural Network for Short-Term Load Forecasting

Short-term load forecasting (STLF) is part and parcel of the efficient working of power grid stations. Accurate forecasts help to detect the fault and enhance grid reliability for organizing sufficient energy transactions. STLF ranges from an hour ahead prediction to a day ahead prediction. Various electric load forecasting methods have been used in literature for electricity generation planning to meet future load demand. A perfect balance regarding generation and utilization is still lacking to avoid extra generation and misusage of electric load. Therefore, this paper utilizes Levenberg–Marquardt (LM) based Artificial Neural Network (ANN) technique to forecast the short-term electricity load for smart grids in a much better, more precise, and more accurate manner. For proper load forecasting, we take the most critical weather parameters along with historical load data in the form of time series grouped into seasons, i.e., winter and summer. Further, the presented model deals with each season’s load data by splitting it into weekdays and weekends. The historical load data of three years have been used to forecast week-ahead and day-ahead load demand after every thirty minutes making load forecast for a very short period. The proposed model is optimized using the Levenberg-Marquardt backpropagation algorithm to achieve results with comparable statistics. Mean Absolute Percent Error (MAPE), Root Mean Squared Error (RMSE), R2, and R are used to evaluate the model. Compared with other recent machine learning-based mechanisms, our model presents the best experimental results with MAPE and R2 scores of 1.3 and 0.99, respectively. The results prove that the proposed LM-based ANN model performs much better in accuracy and has the lowest error rates as compared to existing work.

A Weakly Supervised Transfer Learning Approach for Radar Sounder Data Segmentation

Airborne Radar Sounders (RSs) are active sensors that acquire subsurface data for Earth observation. RS data (radargrams) provide information on buried geology by imaging subsurface dielectric discontinuities. Recently, several automatic RS target identification techniques have been proposed, being convolutional neural network (CNN)-based methods the most promising. However, they require numerous labeled data that are hard to retrieve in the subsurface environment targeted by RS. Further, they are not designed to effectively deal with problems showing unbalanced classes like RS segmentation. We introduce newer cryosphere subsurface targets in the inland and coastal areas that can have a very low probability. To deal with the higher complexity and variability than previous works, we propose a transfer learning framework for RS data to mitigate the need for a large amount of labeled data and handle extremely unbalanced target classes. Herewith, we propose two transfer learning-based mechanisms for radargram segmentation. The first uses a lightweight architecture whose pre-training is supervised with a large labeled dataset from other domains. The second mechanism uses a deep architecture pre-trained in the RS domain, considering the pretest task of radargram reconstruction. The architectures are modified to deal with the characteristics of RS data and the radargram segmentation task. Finally, both methods are fine-tuned with a few labeled radargrams to learn radargram features useful for segmentation. We reveal experimental results on radargrams acquired in Antarctica by MCoRDS-1 and MCoRDS-3. The results demonstrate the effectiveness of transfer learning for radargram segmentation.

TFAD: TCP flooding attack detection in software-defined networking using proxy-based and machine learning-based mechanisms

TFAD: TCP flooding attack detection in software-defined networking using proxy-based and machine learning-based mechanisms

Dynamic Event-Triggered Integral Sliding Mode Adaptive Optimal Tracking Control for Uncertain Nonlinear Systems

In this paper, we study the event-triggered integral sliding mode optimal tracking problem of nonlinear systems with matched and unmatched disturbances. The goal is to design an adaptive dynamic programming-based sliding-mode controller, which stabilizes the closed-loop system and guarantees the optimal performance of the sliding-mode dynamics. First, in order to remove the effects of the matched uncertainties, an event-triggered sliding mode controller is designed to force the state of the systems on the sliding mode surface without Zeno behavior. Second, another event-triggered controller is designed to suppress unmatched disturbances with a nearly optimal performance while also guaranteeing Zeno-free behavior. Finally, the benefits of the proposed algorithm are shown in comparison to several traditional triggering and learning-based mechanisms.

Automatic COVID-19 detection mechanisms and approaches from medical images: a systematic review.

Since early 2020, Coronavirus Disease 2019 (COVID-19) has spread widely around the world. COVID-19 infects the lungs, leading to breathing difficulties. Early detection of COVID-19 is important for the prevention and treatment of pandemic. Numerous sources of medical images (e.g., Chest X-Rays (CXR), Computed Tomography (CT), and Magnetic Resonance Imaging (MRI)) are regarded as a desirable technique for diagnosing COVID-19 cases. Medical images of coronavirus patients show that the lungs are filled with sticky mucus that prevents them from inhaling. Today, Artificial Intelligence (AI) based algorithms have made a significant shift in the computer aided diagnosis due to their effective feature extraction capabilities. In this survey, a complete and systematic review of the application of Machine Learning (ML) methods for the detection of COVID-19 is presented, focused on works that used medical images. We aimed to evaluate various ML-based techniques in detecting COVID-19 using medical imaging. A total of 26 papers were extracted from ACM, ScienceDirect, Springerlink, Tech Science Press, and IEEExplore. Five different ML categories to review these mechanisms are considered, which are supervised learning-based, deep learning-based, active learning-based, transfer learning-based, and evolutionary learning-based mechanisms. A number of articles are investigated in each group. Also, some directions for further research are discussed to improve the detection of COVID-19 using ML techniques in the future. In most articles, deep learning is used as the ML method. Also, most of the researchers used CXR images to diagnose COVID-19. Most articles reported accuracy of the models to evaluate model performance. The accuracy of the studied models ranged from 0.84 to 0.99. The studies demonstrated the current status of AI techniques in using AI potentials in the fight against COVID-19.

A Survey of Datasets, Preprocessing, Modeling Mechanisms, and Simulation Tools Based on AI for Material Analysis and Discovery.

Research has become increasingly more interdisciplinary over the past few years. Artificial intelligence and its sub-fields have proven valuable for interdisciplinary research applications, especially physical sciences. Recently, machine learning-based mechanisms have been adapted for material science applications, meeting traditional experiments’ challenges in a time and cost-efficient manner. The scientific community focuses on harnessing varying mechanisms to process big data sets extracted from material databases to derive hidden knowledge that can successfully be employed in technical frameworks of material screening, selection, and recommendation. However, a plethora of underlying aspects of the existing material discovery methods needs to be critically assessed to have a precise and collective analysis that can serve as a baseline for various forthcoming material discovery problems. This study presents a comprehensive survey of state-of-the-art benchmark data sets, detailed pre-processing and analysis, appropriate learning model mechanisms, and simulation techniques for material discovery. We believe that such an in-depth analysis of the mentioned aspects provides promising directions to the young interdisciplinary researchers from computing and material science fields. This study will help devise useful modeling in the materials discovery to positively contribute to the material industry, reducing the manual effort involved in the traditional material discovery. Moreover, we also present a detailed analysis of experimental and computation-based artificial intelligence mechanisms suggested by the existing literature.

An Ensemble of Prediction and Learning Mechanism for Improving Accuracy of Anomaly Detection in Network Intrusion Environments

The connectivity of our surrounding objects to the internet plays a tremendous role in our daily lives. Many network applications have been developed in every domain of life, including business, healthcare, smart homes, and smart cities, to name a few. As these network applications provide a wide range of services for large user groups, the network intruders are prone to developing intrusion skills for attack and malicious compliance. Therefore, safeguarding network applications and things connected to the internet has always been a point of interest for researchers. Many studies propose solutions for intrusion detection systems and intrusion prevention systems. Network communities have produced benchmark datasets available for researchers to improve the accuracy of intrusion detection systems. The scientific community has presented data mining and machine learning-based mechanisms to detect intrusion with high classification accuracy. This paper presents an intrusion detection system based on the ensemble of prediction and learning mechanisms to improve anomaly detection accuracy in a network intrusion environment. The learning mechanism is based on automated machine learning, and the prediction model is based on the Kalman filter. Performance analysis of the proposed intrusion detection system is evaluated using publicly available intrusion datasets UNSW-NB15 and CICIDS2017. The proposed model-based intrusion detection accuracy for the UNSW-NB15 dataset is 98.801 percent, and the CICIDS2017 dataset is 97.02 percent. The performance comparison results show that the proposed ensemble model-based intrusion detection significantly improves the intrusion detection accuracy.

Computing the Uncontrollable: Insights from Computational Modelling of Learning and Choice in Depression

The learned helplessness (LH) paradigm, developed in experimental animals, has had great influence on the development of models of mood and anxiety disorders. However, the insights from this paradigm have not always translated straightforwardly into human experimental work. In particular, instrumental contingency learning experiments yielded the contradictory finding of more accurate contingency knowledge in depressed individuals (“depressive realism”: DR). A growing literature involving the application of computational modelling applied to human learning studies suggest that possible sequelae of LH in humans may be manifest in one or more potential candidate reinforcement learning-based mechanisms. This analysis suggests that the DR effect is theoretically and empirically independent of LH, despite both LH and DR being related to depression. Overall, the review highlights recent developments within the complex literature of learning and choice in depression and proposes a variety of testable predictions which may serve to delineate more precisely the presence, and neural basis, of LH-like sequelae across social and non-social learning and choice paradigms in depressed patients.

Learning lean: rhythm of production and the pace of lean implementation

PurposeWhy some assembly factories implement a lean program faster than others is an enduring puzzle. We examine the effect of a fundamental characteristic of every assembly factory—its rhythm of production.Design/methodology/approachWe designed a multi-method study and collected data from a leading global equipment manufacturer that launched a lean program across its factory network. We use quantitative data gathered from internal company documents to test our hypothesis that production rhythm affects the pace of lean implementation. We then analyze qualitative data from interviews and factory visits to derive theoretical explanations for how production rhythm affects lean implementation.FindingsConsistent with our hypothesis, we present evidence that factories with faster production rhythms implement lean faster than those with slower rhythms. This evidence is consistent with learning theories as well as the literature on organizational routines and forms of knowledge. We propose a theory of the relation between rhythm and learning in lean implementation.Research limitations/implicationsThe hitherto unexplored relation between production rhythm and lean implementation raises intriguing questions for scholars and ushers new insights into how organizations learn to implement lean.Practical implicationsOrganizations need to calibrate their expectations for lean implementation pace when their factories have widely different production rhythms and find ways to mitigate any adverse effects slower rhythms may have. Organizations can alleviate the unfavorable context of slower rhythms by inculcating practices in the factory that emulate the learning environment present in faster-paced factories.Originality/valueWe contribute novel quantitative and qualitative evidence that production rhythm affects lean implementation through learning-based mechanisms.

Becoming a better parent: Mice learn sounds that improve a stereotyped maternal behavior

While mothering is often instinctive and stereotyped in species-specific ways, evolution can favor genetically “open” behavior programs that allow experience to shape infant care. Among experience-dependent maternal behavioral mechanisms, sensory learning about infants has been hard to separate from motivational changes arising from sensitization with infants. We developed a paradigm in which sensory learning of an infant-associated cue improves a stereotypical maternal behavior in female mice. Mice instinctively employed a spatial memory-based strategy when engaged repetitively in a pup search and retrieval task. However, by playing a sound from a T-maze arm to signal where a pup will be delivered for retrieval, mice learned within 7 days and retained for at least 2 weeks the ability to use this specific cue to guide a more efficient search strategy. The motivation to retrieve pups also increased with learning on average, but their correlation did not explain performance at the trial level. Bilaterally silencing auditory cortical activity significantly impaired the utilization of new strategy without changing the motivation to retrieve pups. Finally, motherhood as compared to infant-care experience alone accelerated how quickly the new sensory-based strategy was acquired, suggesting a role for the maternal hormonal state. By rigorously establishing that newly formed sensory associations can improve the performance of a natural maternal behavior, this work facilitates future studies into the neurochemical and circuit mechanisms that mediate novel sensory learning in the maternal context, as well as more learning-based mechanisms of parental behavior in rodents.

GOLDRUSH

Fraud detection rules, written by domain experts, are often employed by financial companies to enhance their machine learning-based mechanisms for accurate detection of fraudulent transactions. Accurate rule writing is a challenging task where domain experts spend significant effort and time. A key observation is that much of this difficulty originates from the fact that experts typically work as "lone rangers" or in isolated groups to define the rules, or work on detecting frauds in one context in isolation from frauds that occur in another context. However, in practice there is a lot of commonality in what different experts are trying to achieve. In this demo, we present the GOLDRUSH system, which facilitates knowledge sharing via effective adaptation of fraud detection rules from one context to another. GOLDRUSH abstracts the possible semantic interpretations of each of the conditions in the rules in one context and adapts them to the target context. Efficient algorithms are used to identify the most effective rule adaptations w.r.t a given cost-benefit metric. We showcase GOLDRUSH through a reenactment of a real-life fraud detection event. Our demonstration will engage the VLDB'18 audience, allowing them to play the role of experts collaborating in the fight against financial frauds.

Self-organizing neural integration of pose-motion features for human action recognition.

The visual recognition of complex, articulated human movements is fundamental for a wide range of artificial systems oriented toward human-robot communication, action classification, and action-driven perception. These challenging tasks may generally involve the processing of a huge amount of visual information and learning-based mechanisms for generalizing a set of training actions and classifying new samples. To operate in natural environments, a crucial property is the efficient and robust recognition of actions, also under noisy conditions caused by, for instance, systematic sensor errors and temporarily occluded persons. Studies of the mammalian visual system and its outperforming ability to process biological motion information suggest separate neural pathways for the distinct processing of pose and motion features at multiple levels and the subsequent integration of these visual cues for action perception. We present a neurobiologically-motivated approach to achieve noise-tolerant action recognition in real time. Our model consists of self-organizing Growing When Required (GWR) networks that obtain progressively generalized representations of sensory inputs and learn inherent spatio-temporal dependencies. During the training, the GWR networks dynamically change their topological structure to better match the input space. We first extract pose and motion features from video sequences and then cluster actions in terms of prototypical pose-motion trajectories. Multi-cue trajectories from matching action frames are subsequently combined to provide action dynamics in the joint feature space. Reported experiments show that our approach outperforms previous results on a dataset of full-body actions captured with a depth sensor, and ranks among the best results for a public benchmark of domestic daily actions.

The development of an attentional bias for angry faces following Pavlovian fear conditioning

Although it is well documented that fear responses develop following aversive Pavlovian conditioning, it is unclear whether fear learning also manifests in the form of attentional biases for fear-related stimuli. Boschen, Parker, and Neumann (Boschen, M. J., Parker, I., & Neumann, D. L. (2007). Changes in implicit associations do not occur simultaneously to Pavlovian conditioning of physiological anxiety responses. Journal of Anxiety Disorders, 21, 788–803.) showed that despite the acquisition of differential skin conductance conditioned responses to angry faces paired (CS+) and unpaired (CS−) with an aversive shock, development of implicit associations was not subsequently observed on the Implicit Association Test. In the present study, participants ( N = 76) were assigned either to a Shock or NoShock group and completed a similar aversive Pavlovian conditioning procedure with angry face CS+ and CS− stimuli. Participants next completed a visual probe task in which the angry face CS+ and CS− stimuli were paired with angry face control stimuli and neutral faces. Results confirmed that differential fear conditioning was observed in the Shock group but not in the NoShock group, and that the Shock group subsequently showed a selective attentional bias for the angry face CS+ compared with the CS− and control stimuli during the visual probe task. The findings confirm the interplay between learning-based mechanisms and cognitive processes, such as attentional biases, in models of fear acquisition and have implications for treatment of the anxiety disorders.

Learning-based Animal Models: Task-specific Focal Hand Dystonia

Dystonia is a disabling, involuntary disorder of movement that leads to writhing, twisting end-range movements or abnormal postures. Inadequate inhibition could account for excessive excitation and near synchronous co-contractions of agonists and antagonists. Dystonia may be generalized or specific, affecting only one part of the body or involving only a well-learned task (e.g., writing, keyboarding, golfing, playing a musical instrument). Task-specific and other focal dystonias are considered idiopathic, with multiple factors such as genetics, anatomy, physiology, psychology, environment, and behavioral characteristics contributing to the development of symptoms. This article provides detailed descriptions of two behavioral animal models (a primate [owl monkey] model and a rodent [Sprague-Dawley rat] model) developed to study the effect of excessive repetition as a potential etiology of focal hand dystonia (FHd). The hypothesis is that repetitive, near simultaneous hand movements can degrade the topographic representations of the hand on the somatic sensory and motor cortices, creating the involuntary movements characteristic of dystonia. While animal studies permit the opportunity for greater control to determine efficacy, the findings must always be confirmed by clinical studies to evaluate sensitivity and specificity of diagnosis and effectiveness of treatment in the home, work, and personal environment. This article presents a review of the etiology and clinical implications for intervention strategies from animal and clinical studies that support learning-based mechanisms for FHd. Other animal models are also briefly reviewed.