Samar Wazir

Samar Wazir, Gautam Siddharth Kashyap, Parag Saxena

Recently, Machine Learning (ML) has become a widely accepted method for significant progress that is rapidly evolving. Since it employs computational methods to teach machines and produce acceptable answers. The significance of the Machine Learning Operations (MLOps) methods, which can provide acceptable answers for such problems, is examined in this study. To assist in the creation of software that is simple to use, the authors research MLOps methods. To choose the best tool structure for certain projects, the authors also assess the features and operability of various MLOps methods. A total of 22 papers were assessed that attempted to apply the MLOps idea. Finally, the authors admit the scarcity of fully effective MLOps methods based on which advancements can self-regulate by limiting human engagement.

Gautam Siddharth Kashyap, Jatin Sohlot, Ayesha Siddiqui et al.

A health crisis is raging all over the world with the rapid transmission of the novel-coronavirus disease (Covid-19). Out of the guidelines issued by the World Health Organisation (WHO) to protect us against Covid-19, wearing a facemask is the most effective. Many countries have necessitated the wearing of face masks, but monitoring a large number of people to ensure that they are wearing masks in a crowded place is a challenging task in itself. The novel-coronavirus disease (Covid-19) has already affected our day-to-day life as well as world trade movements. By the end of April 2021, the world has recorded 144,358,956 confirmed cases of novel-coronavirus disease (Covid-19) including 3,066,113 deaths according to the world health organization (WHO). These increasing numbers motivate automated techniques for the detection of a facemask in real-time scenarios for the prevention of Covid-19. We propose a technique using deep learning that works for single and multiple people in a frame recorded via webcam in still or in motion. We have also experimented with our approach in night light. The accuracy of our model is good compared to the other approaches in the literature; ranging from 74% for multiple people in a nightlight to 99% for a single person in daylight.

Gautam Siddharth Kashyap, Md Tabrez Nafis, Samar Wazir

Training Artificial Neural Networks (ANNs) with Stochastic Gradient Descent (SGD) frequently encounters difficulties, including substantial computing expense and the risk of converging to local optima, attributable to its dependence on partial weight gradients. Therefore, this work investigates Particle Swarm Optimization (PSO) and Genetic Algorithms (GAs) - two population-based Metaheuristic Optimizers (MHOs) - as alternatives to SGD to mitigate these constraints. A hybrid PSO-SGD strategy is developed to improve local search efficiency. The findings indicate that the hybrid PSO-SGD technique decreases the median training MSE by 90 to 95 percent relative to conventional GA and PSO across various network sizes (e.g., from around 0.02 to approximately 0.001 in the Sphere function). RMHC attains substantial enhancements, reducing MSE by roughly 85 to 90 percent compared to GA. Simultaneously, RS consistently exhibits errors exceeding 0.3, signifying subpar performance. These findings underscore that hybrid and evolutionary procedures significantly improve training efficiency and accuracy compared to conventional optimization methods and imply that the Building Block Hypothesis (BBH) may still be valid, indicating that advantageous weight structures are retained during evolutionary search.