Abstract

In this chapter, we introduce Hopfield neural networks based on the ceiling neuron model. Inspired by the binary Hopfield neural network and by the definition of ceiling neuron, we propose an extension of the Hopfield neural network from a binary set of possible states of a real-valued neuron to a multistate set. We investigate the dynamics of the proposed model in asynchronous and synchronous update modes. Computational experiments illustrate that the Hopfield neural network based on the ceiling neuron in an asynchronous update mode always settles down to a stationary state under the usual conditions on synaptic weights, that is, the synaptic weight matrix is symmetric with non-negative diagonal elements. In turn, the Hopfield neural network with ceiling neurons in the synchronous update mode can generate limit cycles, similarly to the classic Hopfield neural network.

Abstract

In this research paper storage as well as retrieval of 1-D/2-D/3-D information using Hopfield type Associative Memories (AMs) is discussed. Various Artificial Neural Networks(ANN) architectures are proposed. Also, implementational issues associated with those Associative Memories are discussed. Cascade connection of AM and Convolutional Neural Network is proposed for noise immunity. Keywords: Associative Memories, Hopfield Network, Ceiling Neuron

Abstract

Trellis coded modulation (TCM) is a technique combining modulation with coding using trellises designed with heuristic techniques that maximize the minimum Euclidean distance of a codebook. We propose a neural networks based decoder for decoding TCM. We show experiments with our decoder that suggest the use of Convolutional Neural Network (CNN) with Recurrent Neural Network (RNN) can improve decoding performance and provide justification for the same. We show the generalization capability of the decoder by training it with small block length and testing for larger block length. We also test our decoder for its performance on noise model unseen in the training. Index Terms—Trellis Coded Modulation, Neural Networks, Decoding

Abstract

— This paper presents a model predictive control (MPC) based algorithm for tracking multiple targets using a swarm of unmanned aerial vehicles (UAVs). All the UAVs belong to fixed-wing category with constraints on flight velocity, climb rate and turn rate. Each UAV carries a camera to detect and track the target. Two cases are considered where for the first case, the number of the UAVs is equal to the number of targets. For the second case, the number of UAVs is lesser than the number of targets leading to a conservative solution where the objective is to maximize the average time duration for which the targets are in the field-of-view (FOV) of any one of the UAV’s camera. A data driven Gaussian process (GP) based model is developed to relate the hyperparameters used in MPC to the mission efficiency. Bayesian optimization is performed to obtain the hyperparameters of the MPC that maximize the mission efficiency. Numerical simulations are performed for both cases using algorithm based on distributed MPC formulation. A performance comparison is provided with the centralized MPC formulation.

Abstract

Online abuse and offensive language on social media have become widespread problems in today’s digital age. In this paper, we contribute a Reddit-based dataset, consisting of 68,159 insults and 51,102 compliments targeted at individuals instead of targeting a particular community or race. Secondly, we benchmark multiple existing state-of-the-art models for both classification and unsupervised style transfer on the dataset. Finally, we analyse the experimental results and conclude that the transfer task is challenging, requiring the models to understand the high degree of creativity exhibited in the data.

Abstract

Clickbait headlines are compelling and designed to lure readers to click on the linked article. To understand their impact on the credibility of news items, we designed and conducted an experiment on news articles from the WebisClickbait-17 dataset. The role of readers’ age and epistemic curiosity on clickbait propensity was also studied. Two datasets of 6 articles each of news relevant to North America and India were selected and each article was assigned a clickbait and a non-clickbait headline based on a pre-test for clickbait perception. 200 participants (100 each of English-speaking American and English-speaking Indian) were instructed to click and read through any two articles of their choice and fill a credibility questionnaire. Epistemic curiosity and demographic data were collected at the end of the experiment. Clickbait headlines significantly reduced the credibility of news items. A correlation between age and propensity for clickbait was observed. No correlation was observed between diversive epistemic curiosity and clickbait propensity, while a weakly negative correlation was observed between specific epistemic curiosity and clickbait propensity. We hope that our findings may lead to a better end-user experience in terms of access to credible information for news readers, bereft of clickbait. Keywords— clickbait, credibility, journalism, digital media, social

Abstract

Over the past two decades, a growing thrust area of research has been electrification of different mechanical and hydraulic systems in ground vehicles, which offers less environmental concerns due to the removal of hydraulic fluids and continual engine parasitic losses. The electrical equivalents for traditional mechanical linkages and hydraulic power assist systems include ‘brakeby-wire’, ‘steer-by-wire’ and ‘throttle-by-wire’ system. These equivalents, collectively known as ‘X-by-Wire’ technologies, form the fundamental structure for the Semi and Fully Autonomous Driving Systems 1 . As compared to a traditional steering system, a Steer-by-Wire (SBW) system removes the mechanical shaft that connects the steering column and steering pinion: therefore, in absence of any physical connection, the steering input is transmitted via a by-wire electronic communication module 2 (cf. Fig. 1). In order to have the SBW equipped vehicle have the similar features to that of a conventional steering system, it is important to devise a tracking control mechanism for the steering actuator such that the steering commands from hand-wheel are followed as precisely as possible. Initial attempts were made to solve such control problem employing conventional PID controller 3,4; however, in presence of inevitable modelling imprecision and parametric uncertainties, advanced control strategies are eventually

Abstract

Wireless Sensors that are small in size and limited battery capacity are widely used for several different applications. For this reason energy efficient as well as cost of the overall network are the two main factors influencing the Wireless Sensor Networks (WSN) performance. In this research paper, in Wireless Sensor Network, energy efficient placement of sensors on the uniform rectangular grid (as sensor field) is proposed. Also, distributed computation of minimum and/or maximum over such sensor field is discussed (for minimizing computations and communications delay). Index Terms—Sensor Placement, Leveling, Clustering, Energy Efficiency, Network Lifetime, Cost Minimization,

Abstract

In this paper, we quantitatively show the energy savings obtained in a heterogeneous wireless sensor network, by using Weighted Centriod algorithm. Due to the energy constraints on the sensor nodes in hierarchal networks efficient cluster head selection has become crucial factor. We propose an algorithm, in which all the nodes do not necessarily have the same initial energy. Our main contribution is to derive the location of a weighted Centroid which minimizes weights sum of squares distances from the sensor nodes to the cluster head location. Analytical results are validated through simulation in OMNET++ by using various performance metrics and compared with different cluster head selection protocols such as I-Leach, Leach-E and NR-Leach. This shows the promising potentials of our Weighted Centriod algorithm.

Abstract

: Reference Evapotranspiration (ET0) is a complex hydrological variable defined by various climatic variables affecting water and energy balances and critical factors for crop water requirements and irrigation scheduling. Conventionally ET0 is calculated by various empirical methods based on rigorous climatic data. However, there are many places where various climatic data may not available for ET0 estimation. The objective of this study is to evaluate different machine learning (ML) techniques to estimate ET0 with minimal climatic inputs. In this study, FAO56 Penman-Monteith model was considered as the standard model and different ML models based on A Long short-term memory neural networks (LSTM), Gradient Boosting Regressor (GBR), Random Forest (RF) and Support Vector Regression (SVR) were developed to estimate ET0 with climatic variables as input parameters. These models were evaluated in two different climatic regions, Hyderabad in India and Waipara in New Zealand. The results indicated that 99 % accuracy could be achieved with all climatic input, whereas accuracy drops to 86% with fewer data. LSTM model performed better than other ML models with all input combinations at both the stations, followed by SVR and RF. Both LSTM and SVR models have been noted as the most robust ML models for estimating ET0 with minimal climate data. Even though the excellent performance can be achievable when all input variables are used, the study, however, found that even a three-parameter combination (Temperature, Wind Speed and Relative Humidity values) or two-parameter combination (Temperature and Relative Humidity, Temperature and Wind Speed) can also be promising in ET0 estimation. The presented study will help to estimate ET0 for data scare regions, which is vital for agricultural water management in semi-arid climates.