Read, summarize, and present the key ideas, contributions, and
limitations of the paper assigned to you.
Total Marks: 05 Presentation Date: 8th December 2024
Roll No. Topic Description Research Paper
1 Explainable AI (XAI) Techniques for making ML models Enhancing Transparency and Understanding in AI Decision-
interpretable and transparent. Making Processes.
2 Transfer Learning Applications and challenges in transferring Multi-source weighted source-free domain transfer method
knowledge between domains. for rotating machinery fault diagnosis.
3 Self-Supervised Learning How models learn representations without The Pursuit of Human Labeling: A New Perspective on
labelled data Unsupervised Learning.
4 Meta-Learning Learning to learn and its applications in Meta-learning Approaches for Few-Shot Learning: A Survey
few-shot learning. of Recent Advances.
5 Federated Learning Privacy-preserving distributed ML. Towards a privacy-preserving distributed machine learning
framework.
6 Bayesian Machine Learning Probabilistic approaches to ML. A personalized probabilistic approach to ovarian cancer
diagnostics.
7 Optimization in ML Advanced optimization techniques for Advanced hyperparameter optimization of deep learning
training deep models. models for wind power prediction.
8 Active Learning Reducing the need for labelled data by Unlabeled data selection for active learning in image
querying the most informative samples. classification.
9 Transformer Models Applications of transformers beyond NLP GiT: Towards Generalist Vision Transformer through Universal
(e.g., vision transformers). Language Interface.
10 Generative Adversarial Applications in image synthesis, data Visual Ship Image Synthesis and Classification Framework
Networks (GANs) augmentation, etc. Based on Attention-DCGAN
11 Graph Neural Networks (GNNs) Applications in social networks, chemistry, An efficient recommendation model based on Knowledge
and recommendation systems. Graph Attention-assisted Network (KGAT-AX)
12 Neural Architecture Search Automating the design of neural networks. A Scalable Neural Network for DSIC Affine Maximizer Auction
(NAS) Design
13 Energy-Efficient Deep Learning Techniques for reducing the computational Machine learning and deep learning: Methods, techniques,
cost of deep learning. applications, challenges, and future research opportunities
14 ML in Healthcare predicting diseases, drug discovery, and Machine Learning in Modeling Disease Trajectory and
personalized medicine. Treatment Outcomes An Emerging Enabler for Model-
Informed Precision Medicine.
15 ML in Climate Science Predicting weather patterns and climate Lake Water Temperature Modeling in an Era of Climate
change impacts. Change Data Sources, Models, and Future Prospects
16 Foundation Models Large-scale pre-trained models like GPT and Challenges of Using Pre-trained Models - the Practitioners’
their societal impact. Perspective
17 Deep Reinforcement Learning Applications in robotics and game playing. GeoBotsVR - A Robotics Learning Game for Beginners with
Hands-on Learning Simulation.
18 Quantum Machine Learning Combining quantum computing with ML Quantum Computing for Climate Resilience and Sustainability
Challenges
19 Zero-Shot and Few-Shot Learning with minimal labelled data. Uniap - Towards universal animal perception in vision via few-
Learning shot learning
20 ML in Education Personalized learning and automated Harnessing AI for Education 4.0 - Drivers of Personalized
grading systems. Learning
21 Multi-Agent Reinforcement Collaboration and competition in multi- Multi-agent systems for modelling self-organization and
Learning agent systems. cooperation processes
22 Safe Reinforcement Learning Ensuring safety in RL applications. A Review of Safe Reinforcement Learning - Methods,
Theories and Applications
