Artificial Intelligence & Machine Learning

Module 1: Deep Learning Optimization

• Mixed-precision training (NVIDIA AMP)

• Model pruning/distillation (Lottery Ticket Hypothesis)

Module 2: Generative Models

• Diffusion models (stable diffusion, DDPM)

• GAN inversion (latent space manipulation)

Module 3: Reinforcement Learning

• Multi-agent systems (MADDPG, QMIX)

• Meta-RL (MAML, PEARL)

Hands-on Labs:

• Training a transformer model from scratch with JAX

• Deploying RL policies in ROS-based simulators