arXiv subjects

Abhishek Sharma

Publications and source records attributed to Abhishek Sharma.

At least 91 records · Page 5Linked to original sources

arXiv2023

iPAL: A Machine Learning Based Smart Healthcare Framework For Automatic Diagnosis Of Attention Deficit/Hyperactivity Disorder (ADHD)

ADHD is a prevalent disorder among the younger population. Standard evaluation techniques currently use evaluation forms, interviews with the patient, and more. However, its symptoms are similar to those of many other disorders like depression, conduct disorder, and oppositional defiant disorder, and these current diagnosis techniques are not very effective. Thus, a sophisticated computing model holds the potential to provide a promising diagnosis solution to this problem. This work attempts to explore methods to diagnose ADHD using combinations of multiple established machine learning techniques like neural networks and SVM models on the ADHD200 dataset and explore the field of neuroscience. In this work, multiclass classification is performed on phenotypic data using an SVM model. The better results have been analyzed on the phenotypic data compared to other supervised learning techniques like Logistic regression, KNN, AdaBoost, etc. In addition, neural networks have been implemented on functional connectivity from the MRI data of a sample of 40 subjects provided to achieve high accuracy without prior knowledge of neuroscience. It is combined with the phenotypic classifier using the ensemble technique to get a binary classifier. It is further trained and tested on 400 out of 824 subjects from the ADHD200 data set and achieved an accuracy of 92.5% for binary classification The training and testing accuracy has been achieved upto 99% using ensemble classifier.

cs.AI↗

arXiv2023

Bipartite Dielectric Huygens' Metasurface for Anomalous Refraction

Huygens' metasurfaces - fundamentally based on Schelkunoff's equivalence principle, Huygens' metasurfaces consist of a two-dimensional array of Huygens' sources formed by co-located orthogonal electric and magnetic dipoles. Such metasurfaces provide electric and magnetic responses to an incoming electromagnetic (EM) wave, leading to unidirectional scattering and 2$π$ phase coverage. We herein report a near-reflectionless coarsely discretized dielectric Huygens' metasurface that performs anomalous refraction, offering a low-loss platform for wave manipulation at high frequencies as compared to their lossy metallic analogue. The coarse discretization dramatically simplifies the design, resulting in a metasurface that is highly efficient, cost-effective, and robust. In this paper, the proposed metasurface comprises two meta-atoms per period and is hence named the bipartite dielectric Huygens' metasurface. Through full-wave simulations at 28 GHz, we show that the proposed metasurface can reroute an incident EM wave from $θ_i=15^{\circ}$ to $θ_t=-44.5^{\circ}$ with very high efficiency: 87% of the scattered power is anomalously transmitted to $θ_t$. Based on our observations, a coarsely discretized dielectric Huygens' metasurface platform can be efficacious to design meta-devices with multifaceted functionalities in different frequency regimes.

physics.app-ph↗

arXiv2023

Learning Visual Representations for Transfer Learning by Suppressing Texture

Recent literature has shown that features obtained from supervised training of CNNs may over-emphasize texture rather than encoding high-level information. In self-supervised learning in particular, texture as a low-level cue may provide shortcuts that prevent the network from learning higher level representations. To address these problems we propose to use classic methods based on anisotropic diffusion to augment training using images with suppressed texture. This simple method helps retain important edge information and suppress texture at the same time. We empirically show that our method achieves state-of-the-art results on object detection and image classification with eight diverse datasets in either supervised or self-supervised learning tasks such as MoCoV2 and Jigsaw. Our method is particularly effective for transfer learning tasks and we observed improved performance on five standard transfer learning datasets. The large improvements (up to 11.49\%) on the Sketch-ImageNet dataset, DTD dataset and additional visual analyses with saliency maps suggest that our approach helps in learning better representations that better transfer.

cs.CV↗

arXiv2023

Congruences for the difference of even and odd number of parts of the cubic and some analogous partition functions

Partitions wherein the even parts appear in two different colours are known as cubic partitions. Recently, Merca introduced and studied the function $A(n)$, which is defined as the difference between the number of cubic partitions of $n$ into an even number of parts and the number of cubic partitions of $n$ into an odd number of parts. In particular, using Smoot's \textsf{RaduRK} Mathematica package, Merca proved the following congruences by finding the exact generating functions of the respective functions. For all $n\ge0$, \begin{align*}A(9n+5)\equiv 0\pmod 3,\\ A(27n+26)\equiv 0\pmod 3. \end{align*} By using generating function manipulations and dissections, da Silva and Sellers proved these congruences and two infinite families of congruences modulo 3 arising from these congruences. In this paper, by employing Ramanujan's theta function identities, we present simplified formulas of the generating functions from which proofs of the congruences of Merca as well as those of da Silva and Sellers follow quite naturally. We also study analogous partition functions wherein multiples of $k$ appear in two different colours, where $k\in\{3,5,7,23\}$.

math.NT↗

arXiv2022

Federated Learning with Privacy-Preserving Ensemble Attention Distillation

Federated Learning (FL) is a machine learning paradigm where many local nodes collaboratively train a central model while keeping the training data decentralized. This is particularly relevant for clinical applications since patient data are usually not allowed to be transferred out of medical facilities, leading to the need for FL. Existing FL methods typically share model parameters or employ co-distillation to address the issue of unbalanced data distribution. However, they also require numerous rounds of synchronized communication and, more importantly, suffer from a privacy leakage risk. We propose a privacy-preserving FL framework leveraging unlabeled public data for one-way offline knowledge distillation in this work. The central model is learned from local knowledge via ensemble attention distillation. Our technique uses decentralized and heterogeneous local data like existing FL approaches, but more importantly, it significantly reduces the risk of privacy leakage. We demonstrate that our method achieves very competitive performance with more robust privacy preservation based on extensive experiments on image classification, segmentation, and reconstruction tasks.

cs.LG↗

arXiv2022

Future-Focused Control Barrier Functions for Autonomous Vehicle Control

In this paper, we introduce a class of future-focused control barrier functions (ff-CBF) aimed at improving traditionally myopic CBF based control design and study their efficacy in the context of an unsignaled four-way intersection crossing problem for collections of both communicating and non-communicating autonomous vehicles. Our novel ff-CBF encodes that vehicles take control actions that avoid collisions predicted under a zero-acceleration policy over an arbitrarily long future time interval. In this sense the ff-CBF defines a virtual barrier, a loosening of which we propose in the form of a relaxed future-focused CBF (rff-CBF) that allows a relaxation of the virtual ff-CBF barrier far from the physical barrier between vehicles. We study the performance of ff-CBF and rff-CBF based controllers on communicating vehicles via a series of simulated trials of the intersection scenario, and in particular highlight how the rff-CBF based controller empirically outperforms a benchmark controller from the literature by improving intersection throughput while preserving safety and feasibility. Finally, we demonstrate our proposed ff-CBF control law on an intersection scenario in the laboratory environment with a collection of 5 non-communicating AION ground rovers.

math.OC↗

arXiv2022

Preserving Privacy in Federated Learning with Ensemble Cross-Domain Knowledge Distillation

Federated Learning (FL) is a machine learning paradigm where local nodes collaboratively train a central model while the training data remains decentralized. Existing FL methods typically share model parameters or employ co-distillation to address the issue of unbalanced data distribution. However, they suffer from communication bottlenecks. More importantly, they risk privacy leakage. In this work, we develop a privacy preserving and communication efficient method in a FL framework with one-shot offline knowledge distillation using unlabeled, cross-domain public data. We propose a quantized and noisy ensemble of local predictions from completely trained local models for stronger privacy guarantees without sacrificing accuracy. Based on extensive experiments on image classification and text classification tasks, we show that our privacy-preserving method outperforms baseline FL algorithms with superior performance in both accuracy and communication efficiency.

cs.CR↗

arXiv2022

Recent results with radiation-tolerant TowerJazz 180 nm MALTA Sensors

To achieve the physics goals of future colliders, it is necessary to develop novel, radiation-hard silicon sensors for their tracking detectors. We target the replacement of hybrid pixel detectors with Depleted Monolithic Active Pixel Sensors (DMAPS) that are radiation-hard, monolithic CMOS sensors. We have designed, manufactured and tested the MALTA series of sensors, which are DMAPS in the 180 nm TowerJazz CMOS imaging technology. MALTA have a pixel pitch well below current hybrid pixel detectors, high time resolution (< 2 ns) and excellent charge collection efficiency across pixel geometries. These sensors have a total silicon thickness of between 50-300 $μ$m, implying reduced material budgets and multiple scattering rates for future detectors which utilize such technology. Furthermore, their monolithic design bypasses the costly stage of bump-bonding in hybrid sensors and can substantially reduce detector costs. This contribution presents the latest results from characterization studies of the MALTA2 sensors, including results demonstrating the radiation tolerance of these sensors.

physics.ins-det↗

arXiv2022

Surya Namaskar: real-time advanced yoga pose recognition and correction for smart healthcare

Nowadays, yoga has gained worldwide attention because of increasing levels of stress in the modern way of life, and there are many ways or resources to learn yoga. The word yoga means a deep connection between the mind and body. Today there is substantial Medical and scientific evidence to show that the very fundamentals of the activity of our brain, our chemistry even our genetic content can be changed by practicing different systems of yoga. Suryanamaskar, also known as salute to the sun, is a yoga practice that combines eight different forms and 12 asanas(4 asana get repeated) devoted to the Hindu Sun God, Surya. Suryanamaskar offers a number of health benefits such as strengthening muscles and helping to control blood sugar levels. Here the Mediapipe Library is used to analyze Surya namaskar situations. Standing is detected in real time with advanced software, as one performs Surya namaskar in front of the camera. The class divider identifies the form as one of the following: Pranamasana, Hasta Padasana, Hasta Uttanasana, Ashwa - Sanchalan asana, Ashtanga Namaskar, Dandasana, or Bhujangasana and Svanasana. Deep learning-based techniques(CNN) are used to develop this model with model accuracy of 98.68 percent and an accuracy score of 0.75 to detect correct yoga (Surya Namaskar ) posture. With this method, the users can practice the desired pose and can check if the pose that the person is doing is correct or not. It will help in doing all the different poses of surya namaskar correctly and increase the efficiency of the yoga practitioner. This paper describes the whole framework which is to be implemented in the model.

cs.CV↗

arXiv2022

A Probabilistic Chemical Programmable Computer

The exponential growth of the power of modern digital computers is based upon the miniaturisation of vast nanoscale arrays of electronic switches, but this will be eventually constrained by fabrication limits and power dissipation. Chemical processes have the potential to scale beyond these limits performing computations through chemical reactions, yet the lack of well-defined programmability limits their scalability and performance. We present a hybrid digitally programmable chemical array as a probabilistic computational machine that uses chemical oscillators partitioned in interconnected cells as a computational substrate. This hybrid architecture performs efficient computation by distributing between chemical and digital domains together with error correction. The efficiency is gained by combining digital with probabilistic chemical logic based on nearest neighbour interactions and hysteresis effects. We demonstrated the implementation of one- and two- dimensional Chemical Cellular Automata and solutions to combinatorial optimization problems.

cs.ET↗

arXiv2022

Joint Symmetry Detection and Shape Matching for Non-Rigid Point Cloud

Despite the success of deep functional maps in non-rigid 3D shape matching, there exists no learning framework that models both self-symmetry and shape matching simultaneously. This is despite the fact that errors due to symmetry mismatch are a major challenge in non-rigid shape matching. In this paper, we propose a novel framework that simultaneously learns both self symmetry as well as a pairwise map between a pair of shapes. Our key idea is to couple a self symmetry map and a pairwise map through a regularization term that provides a joint constraint on both of them, thereby, leading to more accurate maps. We validate our method on several benchmarks where it outperforms many competitive baselines on both tasks.

cs.CV↗

arXiv2022

A Joint Learning Approach for Semi-supervised Neural Topic Modeling

Topic models are some of the most popular ways to represent textual data in an interpret-able manner. Recently, advances in deep generative models, specifically auto-encoding variational Bayes (AEVB), have led to the introduction of unsupervised neural topic models, which leverage deep generative models as opposed to traditional statistics-based topic models. We extend upon these neural topic models by introducing the Label-Indexed Neural Topic Model (LI-NTM), which is, to the extent of our knowledge, the first effective upstream semi-supervised neural topic model. We find that LI-NTM outperforms existing neural topic models in document reconstruction benchmarks, with the most notable results in low labeled data regimes and for data-sets with informative labels; furthermore, our jointly learned classifier outperforms baseline classifiers in ablation studies.

cs.IR↗

arXiv2022

Real-time Recognition of Yoga Poses using computer Vision for Smart Health Care

Nowadays, yoga has become a part of life for many people. Exercises and sports technological assistance is implemented in yoga pose identification. In this work, a self-assistance based yoga posture identification technique is developed, which helps users to perform Yoga with the correction feature in Real-time. The work also presents Yoga-hand mudra (hand gestures) identification. The YOGI dataset has been developed which include 10 Yoga postures with around 400-900 images of each pose and also contain 5 mudras for identification of mudras postures. It contains around 500 images of each mudra. The feature has been extracted by making a skeleton on the body for yoga poses and hand for mudra poses. Two different algorithms have been used for creating a skeleton one for yoga poses and the second for hand mudras. Angles of the joints have been extracted as a features for different machine learning and deep learning models. among all the models XGBoost with RandomSearch CV is most accurate and gives 99.2\% accuracy. The complete design framework is described in the present paper.

cs.CV↗

arXiv2021

On Learning Prediction-Focused Mixtures

Probabilistic models help us encode latent structures that both model the data and are ideally also useful for specific downstream tasks. Among these, mixture models and their time-series counterparts, hidden Markov models, identify discrete components in the data. In this work, we focus on a constrained capacity setting, where we want to learn a model with relatively few components (e.g. for interpretability purposes). To maintain prediction performance, we introduce prediction-focused modeling for mixtures, which automatically selects the dimensions relevant to the prediction task. Our approach identifies relevant signal from the input, outperforms models that are not prediction-focused, and is easy to optimize; we also characterize when prediction-focused modeling can be expected to work.

cs.LG↗

arXiv2021

Learning Canonical Embedding for Non-rigid Shape Matching

This paper provides a novel framework that learns canonical embeddings for non-rigid shape matching. In contrast to prior work in this direction, our framework is trained end-to-end and thus avoids instabilities and constraints associated with the commonly-used Laplace-Beltrami basis or sequential optimization schemes. On multiple datasets, we demonstrate that learning self symmetry maps with a deep functional map projects 3D shapes into a low dimensional canonical embedding that facilitates non-rigid shape correspondence via a simple nearest neighbor search. Our framework outperforms multiple recent learning based methods on FAUST and SHREC benchmarks while being computationally cheaper, data-efficient, and robust.

cs.CV↗

arXiv2021

On the Robustness of Human Pose Estimation

This paper provides a comprehensive and exhaustive study of adversarial attacks on human pose estimation models and the evaluation of their robustness. Besides highlighting the important differences between well-studied classification and human pose-estimation systems w.r.t. adversarial attacks, we also provide deep insights into the design choices of pose-estimation systems to shape future work. We benchmark the robustness of several 2D single person pose-estimation architectures trained on multiple datasets, MPII and COCO. In doing so, we also explore the problem of attacking non-classification networks including regression based networks, which has been virtually unexplored in the past. \par We find that compared to classification and semantic segmentation, human pose estimation architectures are relatively robust to adversarial attacks with the single-step attacks being surprisingly ineffective. Our study shows that the heatmap-based pose-estimation models are notably robust than their direct regression-based systems and that the systems which explicitly model anthropomorphic semantics of human body fare better than their other counterparts. Besides, targeted attacks are more difficult to obtain than un-targeted ones and some body-joints are easier to fool than the others. We present visualizations of universal perturbations to facilitate unprecedented insights into their workings on pose-estimation. Additionally, we show them to generalize well across different networks. Finally we perform a user study about perceptibility of these examples.

cs.CV↗

arXiv2021

Scale Normalized Image Pyramids with AutoFocus for Object Detection

We present an efficient foveal framework to perform object detection. A scale normalized image pyramid (SNIP) is generated that, like human vision, only attends to objects within a fixed size range at different scales. Such a restriction of objects' size during training affords better learning of object-sensitive filters, and therefore, results in better accuracy. However, the use of an image pyramid increases the computational cost. Hence, we propose an efficient spatial sub-sampling scheme which only operates on fixed-size sub-regions likely to contain objects (as object locations are known during training). The resulting approach, referred to as Scale Normalized Image Pyramid with Efficient Resampling or SNIPER, yields up to 3 times speed-up during training. Unfortunately, as object locations are unknown during inference, the entire image pyramid still needs processing. To this end, we adopt a coarse-to-fine approach, and predict the locations and extent of object-like regions which will be processed in successive scales of the image pyramid. Intuitively, it's akin to our active human-vision that first skims over the field-of-view to spot interesting regions for further processing and only recognizes objects at the right resolution. The resulting algorithm is referred to as AutoFocus and results in a 2.5-5 times speed-up during inference when used with SNIP.

cs.CV↗

arXiv2021

Matrix Decomposition on Graphs: A Functional View

We propose a functional view of matrix decomposition problems on graphs such as geometric matrix completion and graph regularized dimensionality reduction. Our unifying framework is based on the key idea that using a reduced basis to represent functions on the product space is sufficient to recover a low rank matrix approximation even from a sparse signal. We validate our framework on several real and synthetic benchmarks (for both problems) where it either outperforms state of the art or achieves competitive results at a fraction of the computational effort of prior work.

cs.LG↗