arXiv subjects

Muhammad Asad

Publications and source records attributed to Muhammad Asad.

24 records · Page 2Linked to original sources

arXiv2022

ECONet: Efficient Convolutional Online Likelihood Network for Scribble-based Interactive Segmentation

Automatic segmentation of lung lesions associated with COVID-19 in CT images requires large amount of annotated volumes. Annotations mandate expert knowledge and are time-intensive to obtain through fully manual segmentation methods. Additionally, lung lesions have large inter-patient variations, with some pathologies having similar visual appearance as healthy lung tissues. This poses a challenge when applying existing semi-automatic interactive segmentation techniques for data labelling. To address these challenges, we propose an efficient convolutional neural networks (CNNs) that can be learned online while the annotator provides scribble-based interaction. To accelerate learning from only the samples labelled through user-interactions, a patch-based approach is used for training the network. Moreover, we use weighted cross-entropy loss to address the class imbalance that may result from user-interactions. During online inference, the learned network is applied to the whole input volume using a fully convolutional approach. We compare our proposed method with state-of-the-art using synthetic scribbles and show that it outperforms existing methods on the task of annotating lung lesions associated with COVID-19, achieving 16% higher Dice score while reducing execution time by 3$\times$ and requiring 9000 lesser scribbles-based labelled voxels. Due to the online learning aspect, our approach adapts quickly to user input, resulting in high quality segmentation labels. Source code for ECONet is available at: https://github.com/masadcv/ECONet-MONAILabel.

eess.IV↗

arXiv2021

Graphene FET on diamond for high-frequency electronics

Transistors operating at high frequencies are the basic building blocks of millimeter-wave communication and sensor systems. The high velocity and mobility of carriers in graphene can open way for ultra-fast group IV transistors with similar or even better performance than can be achieved with III-V based semiconductors. However, the progress of high-speed graphene transistors has been hampered due to fabrication issues, influence of adjacent materials, and self-heating effects. Here, we report a graphene field-effect transistor (FET) on a diamond substrate, with a $f_{max}$ up to 54 GHz for a gate length of 500 nm. The high thermal conductivity of diamond provides an efficient heat-sink, and its relatively high optical-phonon energy improves saturation velocity of carriers in the graphene channel. Moreover, we show that graphene FETs on diamond, with different gate lengths, exhibit excellent scaling behavior. These results indicate that graphene FETs on diamond technology can reach sub-terahertz frequency performance.

cond-mat.mes-hall↗

arXiv2021

Federated Learning Versus Classical Machine Learning: A Convergence Comparison

In the past few decades, machine learning has revolutionized data processing for large scale applications. Simultaneously, increasing privacy threats in trending applications led to the redesign of classical data training models. In particular, classical machine learning involves centralized data training, where the data is gathered, and the entire training process executes at the central server. Despite significant convergence, this training involves several privacy threats on participants' data when shared with the central cloud server. To this end, federated learning has achieved significant importance over distributed data training. In particular, the federated learning allows participants to collaboratively train the local models on local data without revealing their sensitive information to the central cloud server. In this paper, we perform a convergence comparison between classical machine learning and federated learning on two publicly available datasets, namely, logistic-regression-MNIST dataset and image-classification-CIFAR-10 dataset. The simulation results demonstrate that federated learning achieves higher convergence within limited communication rounds while maintaining participants' anonymity. We hope that this research will show the benefits and help federated learning to be implemented widely.

cs.LG↗

arXiv2020

PROPEL: Probabilistic Parametric Regression Loss for Convolutional Neural Networks

In recent years, Convolutional Neural Networks (CNNs) have enabled significant advancements to the state-of-the-art in computer vision. For classification tasks, CNNs have widely employed probabilistic output and have shown the significance of providing additional confidence for predictions. However, such probabilistic methodologies are not widely applicable for addressing regression problems using CNNs, as regression involves learning unconstrained continuous and, in many cases, multi-variate target variables. We propose a PRObabilistic Parametric rEgression Loss (PROPEL) that facilitates CNNs to learn parameters of probability distributions for addressing probabilistic regression problems. PROPEL is fully differentiable and, hence, can be easily incorporated for end-to-end training of existing CNN regression architectures using existing optimization algorithms. The proposed method is flexible as it enables learning complex unconstrained probabilities while being generalizable to higher dimensional multi-variate regression problems. We utilize a PROPEL-based CNN to address the problem of learning hand and head orientation from uncalibrated color images. Our experimental validation and comparison with existing CNN regression loss functions show that PROPEL improves the accuracy of a CNN by enabling probabilistic regression, while significantly reducing required model parameters by $10 \times$, resulting in improved generalization as compared to the existing state-of-the-art.

cs.CV↗

arXiv2020

Evaluating the Communication Efficiency in Federated Learning Algorithms

In the era of advanced technologies, mobile devices are equipped with computing and sensing capabilities that gather excessive amounts of data. These amounts of data are suitable for training different learning models. Cooperated with advancements in Deep Learning (DL), these learning models empower numerous useful applications, e.g., image processing, speech recognition, healthcare, vehicular network and many more. Traditionally, Machine Learning (ML) approaches require data to be centralised in cloud-based data-centres. However, this data is often large in quantity and privacy-sensitive which prevents logging into these data-centres for training the learning models. In turn, this results in critical issues of high latency and communication inefficiency. Recently, in light of new privacy legislations in many countries, the concept of Federated Learning (FL) has been introduced. In FL, mobile users are empowered to learn a global model by aggregating their local models, without sharing the privacy-sensitive data. Usually, these mobile users have slow network connections to the data-centre where the global model is maintained. Moreover, in a complex and large scale network, heterogeneous devices that have various energy constraints are involved. This raises the challenge of communication cost when implementing FL at large scale. To this end, in this research, we begin with the fundamentals of FL, and then, we highlight the recent FL algorithms and evaluate their communication efficiency with detailed comparisons. Furthermore, we propose a set of solutions to alleviate the existing FL problems both from communication perspective and privacy perspective.

cs.LG↗

arXiv2019

Does carrier velocity saturation help to enhance fmax in graphene field-effect transistors?

It has been argued that current saturation in graphene field-effect transistors (GFETs) is needed to get the highest possible maximum oscillation frequency (fmax). This paper numerically investigates whether velocity saturation can help to get better current saturation and if that correlates with enhanced fmax. For such a purpose, we used a drift-diffusion simulator that includes several factors that influence output conductance, especially at short channel lengths and-or large drain bias: short-channel electrostatics, saturation velocity, graphene-dielectric interface traps, and self-heating effects. As a testbed for our investigation, we analyzed fabricated GFETs with high extrinsin cutoff frequency fT,x (34 GHz) and fmax (37 GHz). Our simulations allow for a microscopic (local) analysis of the channel parameteres such as carrier concentration, drift and saturation velocities. For biases far away from the Dirac voltage, where the channel behaves as unipolar, we confirmed that the higher is the drift velocity, as close as possible to the saturation velocity, the greater fmax is. However, the largest fmax is recorded at biases near the crossover between unipolar and bipolar behavior, where it does not hold that the highest drift velocity maximizes fmax. In fact, the position and magnitude of the largest fmax depend on the complex interplay between the carrier concentration and total velocity which, in turn, are impacted by the self-heating. Importantly, this effect was found to severely limit radio-frequency performance, reducing the maximum fmax from around 60 to 40 GHz.

physics.app-ph↗