Single Circuit in V1 Capable of Switching Contexts during Movement Using an Inhibitory Population as a Switch

As animals adapt to their environments, their brains are tasked with processing stimuli in different sensory contexts. Whether these computations are context dependent or independent, they are all implemented in the same neural tissue. A crucial question is what neural architectures can respond flexibly to a range of stimulus conditions and switch between them. This is a particular case of flexible architecture that permits multiple related computations within a single circuit. Here, we address this question in the specific case of the visual system circuitry, focusing on context integration, defined as the integration of feedforward and surround information across visual space. We show that a biologically inspired microcircuit with multiple inhibitory cell types can switch between visual processing of the static context and the moving context. In our model, the VIP population acts as the switch and modulates the visual circuit through a disinhibitory motif. Moreover, the VIP population is efficient, requiring only a relatively small number of neurons to switch contexts. This circuit eliminates noise in videos by using appropriate lateral connections for contextual spatiotemporal surround modulation, having superior denoising performance compared to circuits where only one context is learned. Our findings shed light on a minimally complex architecture that is capable of switching between two naturalistic contexts using few switching units.

EvoMBN: Evolving Multi-Branch Networks on Myocardial Infarction Diagnosis Using 12-Lead Electrocardiograms

Multi-branch Networks (MBNs) have been successfully applied to myocardial infarction (MI) diagnosis using 12-lead electrocardiograms. However, most existing MBNs share a fixed architecture. The absence of architecture optimization has become a significant obstacle to a more accurate diagnosis for these MBNs. In this paper, an evolving neural network named EvoMBN is proposed for MI diagnosis. It utilizes a genetic algorithm (GA) to automatically learn the optimal MBN architectures. A novel fixed-length encoding method is proposed to represent each architecture. In addition, the crossover, mutation, selection, and fitness evaluation of the GA are defined to ensure the architecture can be optimized through evolutional iterations. A novel Lead Squeeze and Excitation (LSE) block is designed to summarize features from all the branch networks. It consists of a fully-connected layer and an LSE mechanism that assigns weights to different leads. Five-fold inter-patient cross validation experiments on MI detection and localization are performed using the PTB diagnostic database. Moreover, the model architecture learned from the PTB database is transferred to the PTB-XL database without any changes. Compared with existing studies, our EvoMBN shows superior generalization and the efficiency of its flexible architecture is suitable for auxiliary MI diagnosis in real-world.

Development of Advanced Advisory System for Anomalies (AAA) to Predict and Detect the Abnormal Operation in Fired Heaters for Real Time Process Safety and Optimization

As the technology is emerging, the process industries are actively migrating to Industry 4.0 to optimize energy, production, profit, and the quality of products. It should be noted that real-time process monitoring is the area where most of the energies are being placed for the sake of optimization and safety. Big data and knowledge-based platforms are receiving much attention to provide a comprehensive decision support system. In this study, the Advanced Advisory system for Anomalies (AAA) is developed to predict and detect the abnormal operation in fired heaters for real-time process safety and optimization in a petrochemical plant. This system predicts and raises an alarm for future problems and detects and diagnoses abnormal conditions using root cause analysis (RCA), using the combination of FMEA (failure mode and effects analysis) and FTA (fault tree analysis) techniques. The developed AAA system has been integrated with databases in a petrochemical plant, and the results have been validated well by testing the application over an extensive period. This AAA online system provides a flexible architecture, and it can also be integrated into other systems or databases available at different levels in a plant. This automated AAA platform continuously monitors the operation, checks the dynamic conditions configured in it, and raises an alarm if the statistics exceed their control thresholds. Moreover, the effect of heaters’ abnormal conditions on efficiency and other KPIs (key performance indicators) is studied to explore the scope of improvement in heaters’ operation.

Broadband Spectral Analysis Algorithm with High-Frequency Resolution for Elimination of Overlap Interference between Adjacent Channels

Spectral lines can be analysed to determine the physical properties of molecular clouds and the astrochemical processes in the formation area of massive stars. To improve the observation technology of radio astronomy, this paper proposes and compares two spectral analysis algorithms (improved weighted overlap-add (IWOLA) + FFT and IWOLA + weighted overlap-add (WOLA)). The proposed algorithms can obtain an ultra-high-frequency resolution for real-valued wideband signals, eliminate the signal overlapping interference between adjacent channels, substantially decrease the required hardware resources, especially multipliers, adders, and memory resources, and reduce the system design complexity. The IWOLA + FFT algorithm consists of an improved weighted overlap-add (IWOLA) filter bank, fast Fourier transform (FFT), a specific decimation for the output data from the IWOLA filter bank, and a selection for part of the output data from the FFT. The IWOLA + WOLA algorithm consists of the same modules as the IWOLA + FFT algorithm, with the second-stage FFT replaced by the modules of the weighted overlap-add (WOLA) filter bank and the accumulation for each sub-band. Based on an analysis of the underlying principles and characteristics of both algorithms, the IWOLA + FFT algorithm demonstrated a spectrum with a high frequency resolution and a comparable performance to an ultra-large-scale FFT, based on two smaller FFTs and a flexible architecture instead of a ultra-large-scale FFT. The IWOLA + WOLA algorithm contains the same function as the IWOLA + FFT algorithm and demonstrates a higher performance. The proposed algorithms eliminated the interference between the adjacent channels within the entire Nyquist frequency bandwidth. The simulation results verify the accuracy and spectral analysis performances of the proposed algorithms.

The Hanoak House as a Flexible and Adaptable Vernacular Precedent for Modern Architecture

Globally the adverse effects of climate change necessitate the implementation of resilient systems that respond to escalating weather fluctuations and increased urban vulnerability. This requires a shift from the traditional efficiency-focused solutions, towards robust, responsive and flexible models. While novel technologies are being developed to address these needs; existing vernacular examples also present innovative solutions. The purpose of this study is to analyse vernacular solutions, in this case Korean Hanoak housing typologies, in terms their integration of flexible and adaptable spatial and technological systems to inform modern applications. As research method, the study firstly employed an unstructured observational method to document the spatial and technological elements of these vernacular precedents, followed by an intersubjective literature review of these precedents to understand the historic context. As main conclusion the study identified seven design principles to inform the development of flexible and adaptable modern architecture solutions. These include: holistic, integrative design; articulated and reciprocally layered systems; nested levels of flexible and inflexible systems; appropriate scale identification; and appropriate technology use. As contribution, this article analyses existing vernacular precedents and highlights principles that can be applied in various contexts to develop locally responsive and flexible architecture.

A Blended Grammar Learning System Featuring Unsupervised Pattern Discovery

Recent developments in cognitive and psycholinguistic research postulate that language learning is essentially the learning of grammatical construc-tions. An important type of grammatical construction with wide-ranging pedagogical implications is grammar patterns as laid out in Pattern Gram-mar. While grammar patterns have seen increasing adoption in language pedagogy, existing applications typically follow a paper-based, teacher-centered approach to instruction, which is known to be less effective in grammar learning than blended, learner-centered approaches. In this paper, we propose a blended learning model that integrates web-based technology with classroom-based instruction to facilitate efficient, personalized grammar learning. We present the design and implementation of a blended grammar learning system that provides customizable learning materials for individual learners by discovering important grammar patterns from corpora in an unsupervised manner. Preliminary evaluation shows that the proposed system achieves an accuracy in pattern discovery comparable to systems that rely on manually precompiled pattern lists and hard-coded rules. With a flexible architecture and an easy-to-use interface, the system can play a key role in the creation of a blended learning environment that can be integrated into a wide range of language learning curricula.

AutoBandit: A Meta Bandit Online Learning System

Recently online multi-armed bandit (MAB) is growing rapidly, as novel problem settings and algorithms motivated by various practical applications are being studied, building on the top of the classic bandit problem. However, identifying the best bandit algorithm from lots of potential candidates for a given application is not only time-consuming but also relying on human expertise, which hinders the practicality of MAB. To alleviate this problem, this paper outlines an intelligent system called AutoBandit, equipped with many out-of-the-box MAB algorithms, for automatically and adaptively choosing the best with suitable hyper-parameters online. It is effective to help a growing application for continuously maximizing cumulative rewards of its whole life-cycle. With a flexible architecture and user-friendly web-based interfaces, it is very convenient for the user to integrate and monitor online bandits in a business system. At the time of publication, AutoBandit has been deployed for various industrial applications.

Categorizing Sexism and Misogyny through Neural Approaches

Sexism, an injustice that subjects women and girls to enormous suffering, manifests in blatant as well as subtle ways. In the wake of growing documentation of experiences of sexism on the web, the automatic categorization of accounts of sexism has the potential to assist social scientists and policymakers in studying and thereby countering sexism. The existing work on sexism classification has certain limitations in terms of the categories of sexism used and/or whether they can co-occur. To the best of our knowledge, this is the first work on the multi-label classification of sexism of any kind(s). 1 We also consider the related task of misogyny classification. While sexism classification is performed on textual accounts describing sexism suffered or observed, misogyny classification is carried out on tweets perpetrating misogyny. We devise a novel neural framework for classifying sexism and misogyny that can combine text representations obtained using models such as Bidirectional Encoder Representations from Transformers with distributional and linguistic word embeddings using a flexible architecture involving recurrent components and optional convolutional ones. Further, we leverage unlabeled accounts of sexism to infuse domain-specific elements into our framework. To evaluate the versatility of our neural approach for tasks pertaining to sexism and misogyny, we experiment with adapting it for misogyny identification. For categorizing sexism, we investigate multiple loss functions and problem transformation techniques to address the multi-label problem formulation. We develop an ensemble approach using a proposed multi-label classification model with potentially overlapping subsets of the category set. Proposed methods outperform several deep-learning as well as traditional machine learning baselines for all three tasks.