ROCS 2025
Research Opportunities in Computer Science
22nd March 2025 (Saturday)
An outreach event hosted by the Department of CSE at IIT Gandhinagar
Join us for Research Opportunities in Computer Science (ROCS) 2025, an one-day hybrid event showcasing research opportunities in computer science through talks, networking opportunities, and a panel discussion. Please register to participate, latest by Thursday, 20th March.
Registration is free but mandatory: it helps us plan the logistics and is also required if you need a certificate of participation.
Please note that a digital certificate of participation will be provided to registered attendees, subject to participation, after the event.
Note: please click on the talk titles to view the abstracts.
| Time | Event | Speakers |
|---|---|---|
| 10:00 AM | Opening Remarks | |
| 10:15 AM - 11:00 AM | Fairness in AI-based Decision Making Geometry and Material Estimation via Photometric Inverse Rendering |
Manisha Padala Shanmuganathan Raman |
| 11:00 AM - 11:30 AM | Chai Break | |
| 11:30 AM - 1:00 PM | Panel Discussion: Future Research Directions in CS | Nishanth Chandran Jyothi Krishnan Shashank Srikant Pankaj Thorat Moderated by: Nipun Batra |
| 1:00 PM - 2:00 PM | Lunch Break | |
| 2:00 PM - 3:30 PM | Resilient by Design Theoretical Computer Science at IIT Gandhinagar |
Udit Bhatia Abhishek Bichhawat Manoj Gupta |
| 3:30 PM onwards | High Tea + Poster Session + Networking |
The event will be held at IIT Gandhinagar, in the Surendra C Sheth Learning Theatre (previously AB1/101), in Academic Block 1.
Please see this page for more details about reaching IITGN and this page for general information about the IIT Gandhinagar campus.
- Sardar Vallabhbhai Patel International Airport (AMD) is an international airport serving the cities of Ahmedabad and Gandhinagar in Gujarat, India and is nearer to our IIT campus.
- Ahmedabad Junction Railway Station (ADI), locally known as Kalupur station is the main terminus which is very well connected with the other parts of the country.
- Geeta Mandir & Paldi are the bus terminals of Ahmedabad city.
Help us spread the word about ROCS 2025! Please share the poster below with your colleagues and on social media.
How do we model, predict, and design systems that can survive and adapt to climate extremes, urban floods, and ecosystem disruptions? This talk presents a computational perspective on resilience, using tools from network science, physics-guided machine learning, and algorithmic heuristics to address complex, interconnected challenges across both engineered infrastructure and natural systems. We model cities, transportation networks, and even ecosystems as dynamic, multi-layered networks, revealing how climate variability and change propagate risks through these systems. Through real-world examples—including flood prediction, critical infrastructure analysis, and ecosystem connectivity under stress—we demonstrate how hybrid AI models that fuse physical principles with data-driven learning enhance interpretability, scalability, and robustness. By combining core ideas from computer science with domain knowledge in climate and environmental systems, this work advances a new paradigm for building actionable, context-aware resilience in a rapidly changing world.
Passive depth sensing in computer vision is gaining traction due to advantages over active sensors, such as lower cost, reduced power consumption, enhanced portability, and greater depth range. However, these benefits come with increased computational complexity. Traditional geometric methods like multi-view stereo and structure from motion achieve high-quality 3D reconstructions but often struggle with capturing surface and sub-surface reflectance properties, leading to incomplete reconstructions. Photometric methods, including Photometric Stereo and Shape from Polarization, offer a solution by enabling finer geometric detail recovery and the joint estimation of shape, material, and reflectance properties. These techniques extend beyond opaque surfaces to include transparent, translucent, and textureless objects. They also facilitate more complete 3D reconstructions by leveraging changes in light direction and density due to refraction within objects—an aspect often overlooked by traditional geometric approaches. Understanding how refraction modifies image acquisition geometry enhances the depth and accuracy of 3D scene analysis, presenting a rich area for future research. In this talk, we shall understand the fresh perspectives on reconstructing the geometry, appearance, and reflectance of complex scenes through photometric inverse rendering. It will offer practical insights for applying these methods in real-world scenarios, paving the way for advancements across various applications such as physics-based rendering, machine vision, and virtual reality.
We will provide an overview of the work done in theoretical computer science at IIT Gandhinagar. We will also discuss opportunities for students interested in pursuing research in this field. Finally, we will conclude the talk with some interesting and non-trivial problems in algorithms.
Existing AI systems achieve near-optimal results for specific performance measures, often obtained at the cost of certain ethical constraints like fairness and privacy. By fairness, we try to quantify the impact an application has on an individual user (Individual Fairness) or a group of users (Group Fairness). We will specifically look at, Group Fairness in Fair Classification – Classification algorithms form the backbone of many decision-making applications and are shown to be biased towards certain demographic groups. Towards this, we will look at neural network based fair classifiers satisfying group fairness and privacy notions.