PI-810: Artificial Intelligence in Drug Discovery		(2 credits)
1.	Introduction to Deep Learning: Definition and characteristics of deep learning, comparison with traditional machine learning approaches, Deep learning architectures, an overview of popular libraries: TensorFlow, Keras, PyTorch, scikit. Activation functions (softmax, sigmoid, ReLU), Loss functions. Handling overfitting and regularization techniques, learning rate tuning, and early stopping.
2.	Convolutional Neural Network (CNN): Architecture and components of CNNs, Convolutional layers, and filters, Role of pooling layers, and feature reduction. Transfer learning: utilizing pre-trained models, fine-tuning, benefits and limitations of transfer learning. CNN applications in image recognition and drug design.
3.	Recurrent Neural Network (RNN): Understanding sequential data, RNN architectures, and working principles. Long Short-Term Memory (LSTM) networks, applications in natural language processing, and time-series data.
4.	Generative Adversarial Networks (GANs): Basic idea and concept of GANs, components of GAN (generator and discriminator), conditional GAN architecture, case studies of GAN for generating novel molecules with desired properties, property prediction, understanding protein-ligand interactions, Challenges and limitations of GANs in drug discovery (mode collapse, overfitting, generalization, etc)
5.	Reinforcement Learning (RL): Introduction to reinforcement learning, Components of RL: Agent, environment, actions, rewards. RL algorithms (Q-Learning, Deep Q Networks (DQNs), etc). Application of RL in drug discovery: drug lead optimization, de novo drug design, etc.
6.	AI applications in Pharmacoinformatics: Sequence-to-sequence models for SMILES generation, RNN for molecule generation, feature engineering for drugs and targets, Drug-target interaction, Case studies of successful drug repurposing, and virtual and high-throughput screening using AI.
7.	Ethical and Regulation in AI-driven drug discovery: Bias and fairness issues, ensuring data privacy and security concerns in drug discovery datasets, transparency and interpretability of AI models, addressing challenges and limitations in deep learning, regulatory challenges, and guidelines, responsible AI practices in the Pharma and Biotech Industry.