Collaboration is at the heart of innovation at the Department of Computer Science and Engineering. The department actively engage with academic institutions, research organizations, industry partners, and government bodies to foster a dynamic ecosystem for knowledge creation, technology transfer, and societal impact.
Special Interest Groups (SIGs)
The SIGs serve as a student-led academic collectives focused on enhancing research culture, knowledge exchange, and collaborative learning within key areas of Computer Science.
The SIG welcomes collaborations with Industry and Academia to foster innovation, research excellence, and technological advancement in the domain of information security.
SIG | Theoretical Computer Science
The Theoretical Computer Science (TCS) SIG unites faculty and students passionate about the foundations of computing. We explore deep and fundamental questions such as: “What can be computed?”, “How efficiently can it be done?”, “How much efficiency can be achieved in different computational models?”, “How to achieve provable efficiency?” The group focuses on bridging the worlds of algorithm design, computational complexity, and lower-bound theory, fostering collaboration that pushes the limits of what computation can achieve.
Our vision is to build a vibrant, collaborative research community that connects diverse areas of theoretical computer science ranging from complexity theory and geometry to dynamic and distributed algorithms.
The SIG aims to:
Core Research Areas (but not limited to):
Computational Complexity Theory with particular emphasis on the area of Algebraic Complexity Theory that mostly deals with polynomial related complexity questions. Just like Turing machines and Boolean circuits in the Boolean domain, here we have analogous algebraic models such as arithmetic circuits, arithmetic branching programs (ABPs), arithmetic formulas etc. Within this landscape lie several long-standing open problems. Two of those are: 1) Polynomial Identity Testing (PIT), 2) Polynomial Factoring. The problem of PIT asks to test whether a given polynomial is identically zero or not. In the polynomial factorization problem, the goal is to find irreducible factors of a given input polynomial. The input polynomial is represented in one of the algebraic models and the output polynomials are also typically expected in the same representation. Both the problems are interconnected. For general arithmetic circuits, it is provably known that PIT along with univariate factoring is equivalent to multivariate factoring, but this is not clear for more restricted models. This area continues to challenge our understanding of efficiency, structure, and the true limits of algebraic computation.
Computational and Discrete Geometry combines geometry, combinatorics, and algorithms to uncover deep structural properties of geometric graph-like systems. At its core, this area studies finite collections of discrete geometric objects such as point configurations, line segments, and simplices and uses their properties to design fast, elegant algorithms for geometric problems. Some interesting questions in this area include 1. How can we design efficient and visually meaningful graph drawings? 2. Can we prove the existence of small hitting sets for various geometric configurations, and if so, how can we find them efficiently? 3.How do geometric ideas connect with topology, and what new insights emerge from this interaction, especially in the realm of topological combinatorics?
In recent decades, the rise of mobile devices and dynamic networks, such as fast-moving VANETs and highly unstable DTNs has created systems where network topology shifts constantly. These environments have made it clear that traditional static graphs simply can’t capture the evolving nature of these networks. Instead, we need graph models that capture the changes in the graph topology. The central goal in this area is to rethink classic graph problems such as shortest paths, dominating sets, matchings, and minimum spanning trees, etc. when the underlying graph itself is dynamic in nature. This gives rise to a new set of fundamental questions such as 1. How to represent a graph with time-varying topology? 2. How the problem definition changes when changes in the graph topology are incorporated? 3. What is the computational complexity incorporated? Defined problem, 4. How efficiently can it be solved? This field blends theory, modelling, and algorithmic creativity to understand computation in a world where the graph is changing with time.
Distributed Algorithms, particularly for distributed graph problems, offer a radically different view of computation where no single entity has the full picture—each node only knows its local neighbourhood, computation is free, and communication is the real cost. This perspective raises fundamental questions about how efficiently classic graph problems can be solved under strict communication limits. Our goal is to explore these challenges across major distributed models such as the LOCAL, CONGEST, and CONGESTED CLIQUE models. Another goal is to investigate algorithms driven by autonomous mobile agents, which gather information through movement rather than message exchange. Together, these directions aim to reveal what distributed systems can accomplish when information is fragmented and communication is limited.
SIG | AI/ML
The AI/ML SIG brings together faculty members, research scholars, and students dedicated to advancing the frontiers of artificial intelligence and machine learning. Our group spans from foundational learning theories and novel algorithms to their practical application in solving complex, real-world challenges. Our collective goal is to pioneer the development of intelligent systems that are robust, ethical, and human-centric, driving innovation across science, industry, and society.
Our vision is to be a nexus of innovation and collaboration, pioneering the development and responsible deployment of intelligent systems to address global challenges.
The SIG aims to:
Core Research Areas (but not limited to):
This research thrust investigates the core principles that enable intelligent behaviour. It focuses on developing novel algorithms, architectures, and theoretical frameworks for machine learning. Work emphasizes deep learning architectures (CNNs, Transformers, GNNs), reinforcement learning for sequential decision-making, statistical learning theory, and optimization methods for large-scale models. The goal is to create more efficient, generalizable, and powerful learning paradigms.
This area focuses on enabling machines to perceive, understand, and interact with the world through sensory data like images, videos, and text. Key research includes object detection, semantic segmentation, and generative models (GANs, Diffusion Models) for image synthesis.
In NLP, we focus on Large Language Models (LLMs), machine translation, sentiment analysis, and semantic understanding, bridging the gap between human language and machine computation through multimodal AI.
Developing the theoretical foundations for search algorithms, encompassing exhaustive and/or metaheuristic approaches for both search, optimization and machine learning applications. Additionally, exploring newly proposed algorithms for parallel computing architectures, making the search algorithms more scalable and suitable for modern multi-core and distributed environments. A significant challenge addressed is the elimination of redundant exploration during the search process, thereby reducing computational overhead and improving convergence faster. This work not only contributes to the theoretical understanding of search heuristics but also has practical implications in optimizing complex problem-solving across various domains in theoretical computer science.
This crucial domain addresses the ethical, social, and technical challenges of deploying AI systems. Research is centered on Explainable AI (XAI) to make black-box models transparent, methods for ensuring fairness and mitigating bias, and developing robust defenses against adversarial attacks. The focus includes privacy-preserving machine learning (e.g., federated learning, differential privacy) and human-in-the-loop systems, ensuring that AI technology is aligned with human values and is safe, reliable, and accountable.
This thrust applies cutting-edge AI/ML techniques to accelerate discovery and solve pressing global issues. Ongoing work focuses on AI-driven computational biology for drug discovery and genomics, predictive modeling for climate change and sustainable energy systems, AI-powered diagnostics in personalized medicine, and developing intelligent systems for FinTech, smart cities, and autonomous transportation. The aim is to leverage AI as a transformative tool for positive societal change and scientific breakthroughs.
Our research scholars are contributing to several frontier domains:
SIG | Computing Systems
The Computing Systems SIG unites faculty members and research scholars working at the intersection of computing systems, intelligent communication networks, and efficient cyber-physical infrastructure. Our collective goal is to advance interdisciplinary research spanning real-time systems, sustainable networking, emerging Internet architectures, VLSI design, and trustworthy AI-driven systems for next-generation applications.
Our vision is to build an integrated ecosystem of researchers and innovators advancing the frontiers of secure, sustainable, and intelligent computing systems.
The SIG aims to:
Core Research Areas (but not limited to):
Research in these thrust focuses on optimization and scheduling of time-critical systems, such as electric vehicles, smart grids, and intelligent transportation systems. Work emphasizes task scheduling, routing, and energy management under dynamic constraints, leveraging mathematical optimization, metaheuristic algorithms (GA, PSO, SA), and Time-Sensitive Networking (TSN) for deterministic communication. Through IoT- and edge-based architectures, the group develops scalable, reliable, and energy-efficient frameworks for distributed cyber-physical environments.
This line of research promotes green and sustainable communication systems by reducing network power consumption without compromising performance. Key areas include energy-aware routing, adaptive link-rate control, AI-driven resource allocation, and renewable-energy-integrated 5G/6G infrastructures. The focus is on creating environmentally responsible and carbon-neutral networking paradigms for the future of digital connectivity.
This domain integrates hardware innovation with intelligent system design, covering low-power and high-performance VLSI, hardware–software co-design, and AI hardware accelerators. Work spans Design Automation and CAD tools, EDA algorithms, Fault Detection and Yield prediction, 3D IC design and packaging, and Quantum or neuromorphic computing— bridging the gap between circuit-level design and intelligent computing.
This newly emerging thrust integrates distributed computing, federated learning, blockchain, and cybersecurity to enable trustworthy, scalable, and intelligent digital ecosystems. Ongoing work focuses on privacy-preserving federated learning for healthcare and financial data, blockchain-based transparency and smart contracts for supply chains, FinTech, and mining, Cyber-Physical Systems, and Digital Twins for predictive analytics and operational efficiency in agriculture, healthcare, and industrial automation.
SIG | Information Security
The Information Security SIG brings together faculty and students engaged in advancing research and innovation in cybersecurity and data protection. The group focuses on designing and developing secure systems that protect information assets across emerging and traditional computing environments. Members of the SIG contribute to theoretical and applied research that spans cryptography, system security, AI-driven defences, and privacy-preserving technologies. Through sponsored research projects, publications in leading venues, and active mentoring of undergraduate, postgraduate, and doctoral students, the SIG has established itself as a vibrant hub of academic excellence and industry relevance. It also organizes workshops, conferences, bootcamps, and awareness campaigns to promote knowledge dissemination and social responsibility in the digital era.
The Information Security SIG actively partners with industry, government agencies, and academic institutions to co-develop secure solutions for real-world challenges. Through joint research projects, capacity-building initiatives, and knowledge exchange programs, the group seeks to advance India’s cybersecurity ecosystem while contributing globally to the evolution of trustworthy computing.
Our vision is to build a multidisciplinary research community that addresses the security challenges of an increasingly interconnected and intelligent world.
The SIG seeks to:
Core Research Areas (but not limited to):
The group conducts research on classical and post-quantum cryptographic primitives, secure key exchange mechanisms, and resilient protocols that remain robust against quantum adversaries. Focus areas include symmetric and asymmetric cryptography, lightweight cryptography for IoT, and secure multiparty computation frameworks.
Artificial intelligence is leveraged for proactive threat detection, anomaly identification, and behavioural analysis in dynamic network environments. Efforts focus on developing intelligent intrusion detection systems, continuous authentication mechanisms, and reinforcement-learning-driven defence frameworks.
This stream explores techniques for securing distributed and resource-constrained environments such as IoT networks, industrial control systems, and cyber-physical infrastructures. Research emphasizes attack surface mapping, intrusion resilience, and secure communication models for smart environments.
Blockchain and distributed ledger technologies are studied as enablers of transparency, immutability, and decentralized trust. Ongoing work includes smart contract security, consensus vulnerabilities, and applications of blockchain in supply chains and data integrity assurance.
This area focuses on detecting and dissecting malicious code using static and dynamic analysis techniques. The group also investigates vulnerability assessment, exploit mitigation, and penetration testing methodologies to strengthen system defences.
Researchers explore watermarking, steganography, and secure content distribution mechanisms for multimedia data. In parallel, Android and drone security form emerging directions aimed at safeguarding mobile ecosystems and aerial platforms against exploitation.
Collaborations
1. Industry Collaborations
The department has partnered with leading technology companies and industrial organizations to translate research into real-world applications. Some key industry collaborations include:
2. Government & Research Agency Collaborations
The department actively partners with government agencies and premier research institutions to develop innovative solutions for national challenges. Key collaborations include:
3. Academic & International Collaborations
The department collaborates with leading Indian and international universities for joint research, faculty exchange programs, and student engagement activities. Some key collaborations include: