Track 1: Foundations of Machine Learning
- Statistical learning theory and generalization bounds
- Optimization methods for deep learning (adaptive optimizers, loss landscapes)
- Dimensionality reduction and manifold learning
- Graphical models, causal inference, and probabilistic reasoning
- Active learning and query strategies
- Transfer, multi-task, and meta-learning
- Learning from noisy, limited, or imbalanced data
- Reinforcement learning theory and bandit algorithms


Track 3: Pattern Recognition & Computer Vision
- Feature extraction, selection, and descriptor learning
- Object detection, segmentation, and tracking
- Face, gesture, and action recognition
- Medical image analysis and computational pathology
- Remote sensing image analysis and Earth observation
- Document analysis and handwriting recognition
- Biometric recognition (fingerprint, iris, voice)
- 3D shape analysis and point cloud processing


Track 5: Applications of ML & Pattern Recognition
- ML for healthcare (diagnosis, drug discovery, genomics)
- Intelligent transportation and autonomous driving
- Natural language processing and speech recognition
- Recommender systems and personalization
- Time-series forecasting (finance, energy, IoT)
- Robotics and embodied AI
- Smart manufacturing and predictive maintenance
- Agriculture, environmental monitoring, and climate science

Track 2: Deep Learning & Generative Models
- Generative AI (diffusion models, VAEs, GANs, flow-based models)
- Large language models and vision-language models
- Transformer architectures and attention variants
- Self-supervised and foundation model pre-training
- Model compression (pruning, quantization, knowledge distillation)
- Neural architecture search and automated deep learning
- Graph neural networks and geometric deep learning
- Representation learning for video, 3D, and multimodal data


Track 4: Responsible & Trustworthy AI
- Fairness, accountability, and transparency in ML models
- Explainable AI (XAI) and interpretability methods
- Robustness against adversarial attacks and out-of-distribution inputs
- Privacy-preserving ML (federated learning, differential privacy)
- AI safety, value alignment, and ethical frameworks
- Uncertainty quantification and reliable predictions
- Bias detection and mitigation in datasets and algorithms
- Regulatory compliance and auditable AI systems


Track 6: Reinforcement Learning & Decision Intelligence
- Deep reinforcement learning algorithms (DQN, PPO, SAC, TD3)
- Multi-agent reinforcement learning and game-theoretic reasoning
- Inverse reinforcement learning and imitation learning
- Hierarchical reinforcement learning and option frameworks
- Offline reinforcement learning and batch RL
- Reinforcement learning from human feedback (RLHF)
- Sequential decision making under uncertainty (POMDPs, bandits)
- Applications of RL in robotics, autonomous driving, recommendation systems, and game AI