scispace - formally typeset
Search or ask a question
Author

Akhilesh Kumar Sharma

Bio: Akhilesh Kumar Sharma is an academic researcher from Manipal University Jaipur. The author has contributed to research in topics: Computer science & Economics. The author has an hindex of 5, co-authored 41 publications receiving 158 citations. Previous affiliations of Akhilesh Kumar Sharma include Manipal University & Indian Institute of Technology Kanpur.

Papers published on a yearly basis

Papers
More filters
Journal ArticleDOI
TL;DR: A cascaded ensembled network that uses an integration of ConvNet and handcrafted features based multi-layer perceptron is proposed in this work and it is demonstrated that accuracy of ensembleled deep learning model is improved to 98.3% from 85.3%.
Abstract: Skin cancer is caused due to unusual development of skin cells and deadly type cancer. Early diagnosis is very significant and can avoid some categories of skin cancers, such as melanoma and focal cell carcinoma. The recognition and the classification of skin malignant growth in the beginning time is expensive and challenging. The deep learning architectures such as recurrent networks and convolutional neural networks (ConvNets) are developed in the past, which are proven appropriate for non-handcrafted extraction of complex features. To additional expand the efficiency of the ConvNet models, a cascaded ensembled network that uses an integration of ConvNet and handcrafted features based multi-layer perceptron is proposed in this work. This offered model utilizes the convolutional neural network model to mine non-handcrafted image features and colour moments and texture features as handcrafted features. It is demonstrated that accuracy of ensembled deep learning model is improved to 98.3% from 85.3% of convolutional neural network model.

129 citations

Book
21 Dec 2013
TL;DR: Several issues are explained and discussed in this book to establish the connection with the mobile adhoc networks in optimized way and to cover different topological and network study.
Abstract: MANET-Performance Optimization And Issues Akhilesh K. Sharma Mobile Ad-hoc networks are used in many places, especially for the military and flood affected areas are having major concern with the mobile adhoc networks, for establishing the connection with the affected areas. There are several issues are explained and discussed in this book to establish the connection in optimized way. Several routing protocols have been proposed for MANET which can be classified as proactive, reactive and hybrid routing protocols. All of these protocols use blind flooding mechanism for the purpose of broadcasting of route request and route reply messages. This Blind flooding leads to a severe broadcast redundancy causing contention and collision in the network. This book will highlight all the points related to these issues. Key Features : 1. Use of NS-2 Simulator for better understanding of the moving environment of moving Nodes. 2. GUI programming & Sample code to build and simulate the nodes protocols and to change the performance related variables. 3. Topics to cover different topological and network study. 4. Techniques like DSDV, CGSR, AODV etc. for demonstration with examples.

82 citations

Journal ArticleDOI
TL;DR: In this paper, two approaches are used to implement classification models, i.e. 3-layer CNN and RNN-LSTM, and SVM (Sigmoid, Polynomial & Gaussian Kernel).
Abstract: Music is a heavenly way of expressing feelings about the world. The language of music has vast diversity. For centuries, people have indulged in debates to stratisfy between Western and Indian Classical Music. But through this paper, an understanding can be fabricated while differentiating the types of Indian Classical Music. Classical music is one of the essential characteristics of Indian Cultural Heritage. Indian Classical Music is divided into two major parts, i.e. Hindustani and Carnatic. Models have been sculptured and trained to classify between Hindustani and Carnatic Music. In this paper, two approaches are used to implement classification models. MFCCs are used as features and implemented models like DNN (1 Layer, 2 Layers, 3 Layers), CNN (1 Layer, 2 Layers, 3 Layers), RNN-LSTM, SVM (Sigmoid, Polynomial & Gaussian Kernel) as one approach. A 3 channels input is created by merging features like MFCC, Spectrogram and Scalogram and implemented models like VGG-16, CNN (1 Layer, 2 Layers, 3 Layers), ResNet-50 as another approach. 3 Layered CNN and RNN-LSTM model performed best among all the approaches.

37 citations

Journal ArticleDOI
TL;DR: In this paper, the authors collected and analyzed temperature, rainfall, soil, seed, crop production, humidity and wind speed data (in a few regions), which will help the farmers improve the produce of their crops.
Abstract: This paper aims at collecting and analysing temperature, rainfall, soil, seed, crop production, humidity and wind speed data (in a few regions), which will help the farmers improve the produce of their crops. Firstly, we pre-process the data in a Python environment and then apply the MapReduce framework, which further analyses and processes the large volume of data. Secondly, k-means clustering is employed on results gained from MapReduce and provides a mean result on the data in terms of accuracy. After that, we use bar graphs and scatter plots to study the relationship between the crop, rainfall, temperature, soil and seed type of two regions (Ahmednagar, Maharashtra and, Andaman and Nicobar Islands). Further, a self-designed recommender system has been used to predict the crops and display them on a Graphic User Interface designed in a Flask environment. The system design is scalable and can be used to find the recommended crops of other states in a similar manner in the future.

20 citations

Journal ArticleDOI
TL;DR: Shamiktiwari et al. as discussed by the authors proposed a ConvNet model trained by Hybrid Constant-Q Transform (HCQT) for heart sound beat classification, which achieved 96% in multi-class classification.
Abstract: A Phonocardiogram (PCG) signal represents murmurs and sounds signals made by vibrations caused for the period of a cardiac cycle. Acoustic wave generated through the beat of the cardiac cycle propagates through the chest wall. It can be easily recorded by a low-cost small handheld digital device called a stethoscope. It provides information like heart rate, intensity, tone, quality, frequency, and location of various components of cardiac sound. Due to these characteristics, phonocardiogram signals can be used to detect heart status at an early stage in a non-invasive manner. In previous studies, the Convolutional Neural Network (ConvNet) is the most studied architecture, which was fed by features, namely Mel Frequency Cepstral (MFC), Chroma Energy Normalized Statistics (CENS), and Constant-Q Transform (CQT). This work has proposed a ConvNet model trained by Hybrid Constant-Q Transform (HCQT) for heart sound beat classification. CQT, Variable-Q Transform (VQT), and HCQT are extracted from each phonocardiogram signal as the acoustic features, including the dominant MFCC features, feed into five-layer regularized ConvNets. After analyzing the literature in the same domain, it can be stated that this is the first time HCQT is being utilized for PCG signals. The findings of the experiments demonstrate that HCQT is more effective than standard CQT and other variants. Also, the accuracies of the system proposed in this work on the validation datasets are 96% in multi-class classification, which outperforms the proposed work relative to other models significantly. The source code is available on the Github repository https://github.com/shamiktiwari/ PCG-signal-Classification-using-Hybrid-Constant-Q-Transform to support the research community.

20 citations


Cited by
More filters
Posted Content
TL;DR: WILDS is presented, a benchmark of in-the-wild distribution shifts spanning diverse data modalities and applications, and is hoped to encourage the development of general-purpose methods that are anchored to real-world distribution shifts and that work well across different applications and problem settings.
Abstract: Distribution shifts -- where the training distribution differs from the test distribution -- can substantially degrade the accuracy of machine learning (ML) systems deployed in the wild. Despite their ubiquity, these real-world distribution shifts are under-represented in the datasets widely used in the ML community today. To address this gap, we present WILDS, a curated collection of 8 benchmark datasets that reflect a diverse range of distribution shifts which naturally arise in real-world applications, such as shifts across hospitals for tumor identification; across camera traps for wildlife monitoring; and across time and location in satellite imaging and poverty mapping. On each dataset, we show that standard training results in substantially lower out-of-distribution than in-distribution performance, and that this gap remains even with models trained by existing methods for handling distribution shifts. This underscores the need for new training methods that produce models which are more robust to the types of distribution shifts that arise in practice. To facilitate method development, we provide an open-source package that automates dataset loading, contains default model architectures and hyperparameters, and standardizes evaluations. Code and leaderboards are available at this https URL.

579 citations

Journal Article
TL;DR: This work presents the first experiences in using PROB on several case studies, highlighting that PROB enables users to uncover errors that are not easily discovered by existing tools.
Abstract: We present PROB, an animation and model checking tool for the B method PROB's animation facilities allow users to gain confidence in their specifications, and unlike the animator provided by the B-Toolkit, the user does not have to guess the right values for the operation arguments or choice variables PROB contains a model checker and a constraint-based checker, both of which can be used to detect various errors in B specifications We present our first experiences in using PROB on several case studies, highlighting that PROB enables users to uncover errors that are not easily discovered by existing tools

541 citations

Posted Content
TL;DR: This paper demonstrates that in such settings, an adversary can perform a training data extraction attack to recover individual training examples by querying the language model, and finds that larger models are more vulnerable than smaller models.
Abstract: It has become common to publish large (billion parameter) language models that have been trained on private datasets. This paper demonstrates that in such settings, an adversary can perform a training data extraction attack to recover individual training examples by querying the language model. We demonstrate our attack on GPT-2, a language model trained on scrapes of the public Internet, and are able to extract hundreds of verbatim text sequences from the model's training data. These extracted examples include (public) personally identifiable information (names, phone numbers, and email addresses), IRC conversations, code, and 128-bit UUIDs. Our attack is possible even though each of the above sequences are included in just one document in the training data. We comprehensively evaluate our extraction attack to understand the factors that contribute to its success. Worryingly, we find that larger models are more vulnerable than smaller models. We conclude by drawing lessons and discussing possible safeguards for training large language models.

496 citations

Journal ArticleDOI
TL;DR: A functional framework is provided that identifies the acquisition, management, processing and mining areas of IoT big data, and several associated technical modules are defined and described in terms of their key characteristics and capabilities.
Abstract: Internet of Things (IoT) related applications have emerged as an important field for both engineers and researchers, reflecting the magnitude and impact of data-related problems to be solved in contemporary business organizations especially in cloud computing. This paper first provides a functional framework that identifies the acquisition, management, processing and mining areas of IoT big data, and several associated technical modules are defined and described in terms of their key characteristics and capabilities. Then current research in IoT application is analyzed, moreover, the challenges and opportunities associated with IoT big data research are identified. We also report a study of critical IoT application publications and research topics based on related academic and industry publications. Finally, some open issues and some typical examples are given under the proposed IoT-related research framework.

456 citations

Journal Article
TL;DR: This work proposes an alternative unsupervised strategy to learn medical visual representations directly from the naturally occurring pairing of images and textual data, and shows that this method leads to image representations that considerably outperform strong baselines in most settings.
Abstract: Learning visual representations of medical images is core to medical image understanding but its progress has been held back by the small size of hand-labeled datasets. Existing work commonly relies on transferring weights from ImageNet pretraining, which is suboptimal due to drastically different image characteristics, or rule-based label extraction from the textual report data paired with medical images, which is inaccurate and hard to generalize. We propose an alternative unsupervised strategy to learn medical visual representations directly from the naturally occurring pairing of images and textual data. Our method of pretraining medical image encoders with the paired text data via a bidirectional contrastive objective between the two modalities is domain-agnostic, and requires no additional expert input. We test our method by transferring our pretrained weights to 4 medical image classification tasks and 2 zero-shot retrieval tasks, and show that our method leads to image representations that considerably outperform strong baselines in most settings. Notably, in all 4 classification tasks, our method requires only 10% as much labeled training data as an ImageNet initialized counterpart to achieve better or comparable performance, demonstrating superior data efficiency.

266 citations