http://repository.iiitd.edu.in/xmlui/handle/123456789/1319 Year-2023 2026-04-11T13:21:05Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1760 Analysis of the impacts of MGNREGA using spatial data analysis Arora, Utkarsh; Arora, Gaurav (Advisor) In response to the escalating global demand for secure access to freshwater resources, governments worldwide have implemented technology-driven initiatives for water resource development. This research focuses on a pivotal initiative, the Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) of 2005, which has been instrumental in constructing water conservation structures and other assets in rural areas to enhance community welfare and generate employment opportunities. Under the program, over 120 million assets have been constructed across India over the course of seventeen years since the initiative was first announced. There exists a knowledge gap regarding the evaluation of the investment returns derived from the construction of water conservation assets, limiting the ability of policymakers in the Ministry of Rural Development (MoRD). This study aims to bridge this gap, by leveraging various geospatial technology and datasets. For the analysis, diverse spatially delineated data sources from across the country and over different years have been collected and integrated. To facilitate this, a robust data warehouse has been developed, utilizing concepts of computer networks and parallelization for dataset extraction, along with natural language processing (NLP) for seamless integration. The study employs a combination of spatial data analysis, econometrics, and machine learning techniques to investigate the factors influencing the construction of Farm Ponds in specific locations, addressing the questions of where, when, why, and how long these constructions occur. Commencing with an in-depth analysis of the state of Uttar Pradesh, the scope of the study expands to encompass the entirety of India. 2023-12-12T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1755 Collaborative and cross-modal distillation for large language models Dixit, Shantanu; Akhtar, Md Shad (Advisor) Knowledge distillation is a technique that involves transferring knowledge from a larger teacher model to a smaller student model. The latest developments in meta-learning-based knowledge distillation emphasize the significance of fine-tuning the teacher models while taking into account the student’s need for better knowledge distillation. Nevertheless, current MetaKD methods frequently fail to provide incentives for the teacher model to improve itself. We introduce a meta-policy distillation technique aiming to foster both collaboration and competition during the fine-tuning of the teacher model within the meta-learning phase. Additionally, we put forth a curriculum learning framework tailored for the student model within a competitive setting. In this context, the student model endeavors to surpass the teacher model through self-training on a diverse range of tasks. We conduct extensive experiments on two NLU benchmarks GLUE andSuperGLUE [74, 75] and validate our methodology’s effectiveness against various KD techniques.As an extension to the above work we further explore the setting where teacher-student modalities differ (ex: text-vision and vice-versa). Existing cross-modal distillation approaches predominantly utilize modality-dependent features for knowledge distillation, and therefore, fail to adaptively learn the abstractions in different modalities. we propose a generic and modality-agnostic cross-modal distillation technique that can distil knowledge from any arbitrary cross-modal open or closed teacher model to any arbitrary student model in any different modality. Our empirical studies encompass eight natural language understanding tasks and an image classification task, showcasing the efficacy of cross-modal distillation in enhancing the performance of student models. 2023-11-29T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1750 Multi-modal fusion transformer for understanding digital advertisements Khurana, Varun; Shah, Rajiv Ratn (Advisor) In today’s world, digital-born media, especially advertisements, have a substantial influence on our daily lives, from persuading us to buy particular brands to creating awareness about a social or environmental cause. This work proposes LearnAd, a learning method for the challenging task of understanding advertisements. Marketing graphics such as advertisements are digitally borne, multi modal (contain both text and visual content) and employ rhetorical devices such as emotions, symbolism, and slogans to convey meaning. On the other hand, most of the work in visual content understanding today is about camera shot images which does not translate well to marketing graphics To address this gap, we propose using human content interaction patterns in the form of eye movements to finetune the understanding of Vision Transformer (ViT). This helps LearnAd – a multimodal transformer-based cross-attention model, achieve state of the art results on three advertisement understanding tasks – generation of the action that an ad persuades a user to take and the reason it provides for the action (what-why of the ad), and prediction of the sentiment and topic of the advertisement image. Despite the lack of availability of real customer gaze patterns over marketing images, LearnAd achieves state of the art performance on three advertisement understanding tasks with the help of generated human saliency patterns. 2023-05-10T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1745 Bridging congestion state in TCP and QUIC Ambooken, Alvin Joseph; Maity, Mukulika (Advisor) Serverless computing has emerged as a transformative field in the cloud computing world. The high energy and cost savings associated with serverless are leading to its increased prevalence.However, network communications within serverless platforms continue to use the traditional TCP, exposing the protocol’s inherent problems in the transport layer, such as latency and performance overheads. In contrast, QUIC provides features such as built-in security and high-speed data transfer, making it a promising solution in serverless. Through our study, we focus on identifying real-world use cases where QUIC performs better than TCP, offering empirical analysis that guides the adoption of QUIC in the face of evolving technological paradigms. 2023-04-29T00:00:00Z