Mr.Naveen Kumar K,Indian Institute of Technoloy Hyderabad (IIT Hyderabad),India
Mr. Naveen Kumar K is affiliated with the Indian Institute of Technology Hyderabad (IIT Hyderabad), located in India. As of the latest update, Mr. Naveen Kumar K’s specific role or title was not provided. Generally, at institutions like IIT Hyderabad, individuals often hold positions related to teaching, research, administration, or technical support. For a more detailed bio, including academic background, research interests, or notable achievements,
Professional Profile:
π Education:
PhD (Computer Science & Engineering),Indian Institute of Technology Hyderabad,Duration: Jan 2020 – Present,Research Area: Security and Privacy for Machine Learning,CGPA: 9.38 out of 10,Supervisor: Prof. C Krishna Moha,MTech (Computer Science & Engineering),Indian Institute of Technology Hyderabad,Duration: Jan 2019 – Dec 2019,Thesis Title: Defining Traffic States Using Spatio Temporal Traffic Graphs on Aerial Videos,CGPA: 8.65 out of 10,Supervisor: Prof. C Krishna Moha,BTech (Computer Science & Engineering), Indian Institute of Information Technology, Vadodara,Duration: July 2014 – May 2018,CGPA: 8.97 out of 10
πΌ Professional Experience:
SahajAI (Bangalore): Optimized defence against poisoning attacks in federated learning for medical image classification (Oct 2023 – Mar 2024),Visiting Research Scholar – University of Agder, Norway: Optimized model poisoning attack in federated learning (Jan 2023 – July 2023),Visiting Research Scholar – Purdue University, USA: Mitigate data poisoning attacks in federated learning using a precision-guided approach (May 2022 – Sep 2022),TCS Research & Innovation Labs, Hyderabad: Non-convex optimization approach to mitigate data poisoning attacks in federated learning (Jan 2022 – Dec 2022),Visiting Research Scholar – Hiroshima University, Japan: Zero-shot 2D object detection in Autonomous Vehicles (Aug 2021 – Nov 2021)
π§ Technical Skills:
- Machine learning, deep learning (supervised and unsupervised), computer vision
- Programming & Libraries: Python, TensorFlow, PyTorch, OpenCV
πAcademic Achievements & Awards:
- PhD Research Excellence Award 2024 (IIT Hyderabad)
- Shortlisted for Google Research Week (2022, 2023)
- Finalist in Nvidia AI Hackathon finals 2019
- Selected for IITH-RU Project-Based Learning Program
Projects:
- Medicine from the sky: AI-based real-time lightweight system for medical drone delivery
- iV4V (Intelligent Voice for Vision): Audio assistance for visual impairment using AI
- M2Smart: Smart Cities project based on sensing, network, and big data analysis
π¬ Research Focus:
Primary Research Interests:
- Computer Vision
- Machine Learning
- Deep Learning
- Federated Learning
- Privacy and Security
- Autonomous Vehicle Technology
- Black-box adversarial attacks in autonomous vehicle technology
- This paper likely explores vulnerabilities in autonomous vehicle systems when subjected to adversarial attacks that manipulate inputs in ways imperceptible to human senses but can mislead AI models.
- Towards a Transitional Weather Scene Recognition Approach for Autonomous Vehicles
- This research aims to improve autonomous vehicles’ ability to recognize and navigate through varying weather conditions, which are critical for safety and reliability.
- The Impact of Adversarial Attacks on Federated Learning: A Survey
- This survey paper discusses the vulnerabilities of federated learning systems to adversarial attacks, which is crucial for understanding security challenges in collaborative machine learning environments.
- Defining traffic states using spatio-temporal traffic graphs
- Focuses on utilizing spatio-temporal traffic graphs to define and analyze different traffic states, which could aid in optimizing traffic management and predicting congestion.
- Open-air Off-street Vehicle Parking Management System Using Deep Neural Networks: A Case Study
- Presents a case study on deploying deep learning techniques for managing open-air off-street vehicle parking, illustrating practical applications of AI in urban infrastructure.
- Precision Guided Approach to Mitigate Data Poisoning Attacks in Federated Learning
- Discusses methods to mitigate data poisoning attacks specifically targeted at federated learning setups, ensuring the integrity and reliability of collaborative machine learning models.
- Towards Smarter Transport: Harnessing AI in Image Processing for Effective Vehicle Counting
- Explores the application of AI in image processing for accurately counting vehicles, which is essential for traffic management and urban planning.
- Revamping Federated Learning Security from a Defender’s Perspective: A Unified Defense with Homomorphic Encrypted Data Space
- Proposes enhanced security measures for federated learning environments, focusing on leveraging homomorphic encryption to protect sensitive data during collaborative model training.