A5030635177- Landslides and related hazards
- Complex Network Analysis Techniques
- Opinion Dynamics and Social Influence
- COVID-19 diagnosis using AI
- Anomaly Detection Techniques and Applications
- Rock Mechanics and Modeling
- Image and Video Quality Assessment
- Graph theory and applications
- Geotechnical Engineering and Analysis
- Image Enhancement Techniques
- Insect Pest Control Strategies
- Industrial Automation and Control Systems
- Artificial Intelligence in Healthcare
- Entomopathogenic Microorganisms in Pest Control
- Radiomics and Machine Learning in Medical Imaging
- Advanced Image Processing Techniques
- Pneumonia and Respiratory Infections
- Cloud Computing and Resource Management
- Video Surveillance and Tracking Methods
- Sentiment Analysis and Opinion Mining
- Advanced Text Analysis Techniques
- Advanced Steganography and Watermarking Techniques
- Electrocatalysts for Energy Conversion
- Generative Adversarial Networks and Image Synthesis
- Machine Learning in Healthcare
Indian Agricultural Research Institute
2023-2025
Central Building Research Institute
2024
Academy of Scientific and Innovative Research
2024
Delhi Technological University
2021-2024
Institute of Life Sciences
2024
Indraprastha Institute of Information Technology Delhi
2023
Dehradun Institute of Technology University
2023
Galgotias University
2023
Gurukul Kangri Vishwavidyalaya
2022
Indian Institute of Technology Jammu
2021-2022
We present Gandivafair, a distributed, fair share scheduler that balances conflicting goals of efficiency and fairness in GPU clusters for deep learning training (DLT). Gandivafair provides performance isolation between users, enabling multiple users to single cluster, thus, maximizing cluster efficiency. is the first allocates cluster-wide time fairly among active users.
Traditional end-host network stacks are struggling to keep up with rapidly increasing datacenter access link bandwidths due their unsustainable CPU overheads. Motivated by this, our community is exploring a multitude of solutions for future stacks: from Linux kernel optimizations partial hardware offload clean-slate userspace specialized host hardware. The design space explored these would benefit detailed understanding inefficiencies in existing stacks.
We propose a two-stage Convolutional Neural Network (CNN) based classification framework for detecting COVID-19 and Community Acquired Pneumonia (CAP) using the chest Computed Tomography (CT) scan images. In first stage, an infection - or CAP, is detected pre-trained DenseNet architecture. Then, in second fine-grained three-way done EfficientNet The proposed COVID+CAP-CNN achieved slice-level accuracy of over 94% at identifying CAP. Further, has potential to be initial screening tool...
Aim: The traditional harvesting of cauliflower has been done by hand using labor-intensive, expensive, and time-consuming techniques that frequently lead to significant yield losses because immature curds are harvested without selection. A rising number people interested in creating selective harvesters can precisely detect gather healthy, mature heads while reducing damage, waste, labor-intensive operations response these difficulties. purpose this research article is present a thorough...
Efficient nitrogen (N) management plays a critical role in optimizing rice yield and ensuring economic sustainability. This study evaluates the impact of variable levels on dynamics viability transplanted (TPR) direct seeded (DSR). Field experiments were conducted research farm ICAR-IARI, New Delhi year 2022 2023 randomised complete block design with seven N treatments (0, 40, 80, 120, 160, 200 240 kg N/ha) applied split application method. Key components like grain straw yields recorded....
Quality assessment of 3D-synthesized images has traditionally been based on detecting specific categories distortions such as stretching, black-holes, blurring, etc. However, approaches have limitations in accurately entirely 3D synthesized affecting their performance. This work proposes an algorithm to efficiently detect the and subsequently evaluate perceptual quality images. The process generation produces a few pixel shift between reference image, hence they are not properly aligned with...
Abstract: Drug discovery and development (DDD) is a highly complex process that necessitates precise monitoring extensive data analysis at each stage. Furthermore, the DDD both timeconsuming costly. To tackle these concerns, artificial intelligence (AI) technology can be used, which facilitates rapid of datasets within limited timeframe. The pathophysiology cancer disease complicated requires research for novel drug development. first stage in involves identifying targets. Cell structure...
Recently fuel cell has been paid significant attention because of their several advantages namely higher efficiency and eco-friendly technology finds applications in vehicles, various house hold applications, electronic devices, to support the electricity grids. In this paper technologies, working principle, are presented. Also mathematical modeling Simulink model Proton Exchange Membrane (PEMFC) Solid Oxide (SOFC), operating characteristics presented details. The effect parameters on PEMFC...
Recent developments and evolutions in machine learning algorithms have directed to the emergence of first-class altered pictures that create video frames a striking similarity real-world frames. This can disastrous effect on how people perceive digitally available knowledge or facts. camera function allowed compose sounds be convincingly altered. These deep-fake films differ from actual videos by replacing soundtrack, lip movement, location where was captured (background). Many relevant...
Let $A$ and $B$ be two nonempty finite sets of integers let $r$ a positive integer. Define $A+r\cdot B:=\{a+rb:a\in A,\, b\in B \}$. In case $A=B$, Freiman et al. proved that $|A+r\cdot A|\geq 4|A|-4$ for $r\geq 3$. For $r=2$, they obtained an exte
This paper introduces Gamified Adversarial Prompting (GAP), a framework that crowd-sources high-quality data for visual instruction tuning of large multimodal models. GAP transforms the collection process into an engaging game, incentivizing players to provide fine-grained, challenging questions and answers target gaps in model's knowledge. Our contributions include (1) approach capture question-answer pairs from humans directly address weaknesses knowledge, (2) method evaluating rewarding...
Abstract This study has performed molecular modeling, vibrational dynamics, and thermodynamics of tyramine neurotransmitter its monohydrate cluster in gas phase. All theoretical calculations for are at DFT/B3LYP/6‐31++G(d,p) level to optimize the electronic structures calculation frequencies. Most frequencies have been found good agreement with experimentally reported IR Raman However, some modified. The presence water monomer decreased highest orbital–lowest unoccupied orbital (HOMO–LUMO)...