Master advanced AI paradigms through reinforcement learning and generative AI. Build intelligent agents, train game-playing AI, create generative mode...
Lessons
18
Activities
18
Projects
7
Duration
24 sessions • 6 months
Master advanced AI paradigms through reinforcement learning and generative AI. Build intelligent agents, train game-playing AI, create generative models, and understand how RL and GenAI converge through RLHF.
This course consists of 24 structured sessions over 6 months. Each session includes lessons, activities, projects, and assessments to ensure comprehensive learning.
Agent-environment interaction, MDP basics
Tabular Q-Learning, epsilon-greedy exploration
Function approximation, experience replay
Train DQN agent to master Atari game
Policy-based RL, REINFORCE algorithm
Actor-critic architecture, advantage estimation
Trust region methods, PPO clipping
PPO agent for continuous control
Bandit problems, UCB, contextual bandits
Debug RL failures, reward shaping
Generative vs discriminative, latent spaces
Encoder-decoder, reparameterization trick
GAN art generation and VAE latent space exploration
Adversarial training, minimax objective
StyleGAN features, conditional GANs, WGAN-GP
Denoising diffusion, U-Net architecture
Self-attention, autoregressive generation
LLM architecture, prompt engineering
3-stage RLHF pipeline, reward model training
Align LLM with human preferences
Cross-modal learning, CLIP, text-to-image
Text-to-image and image-to-text pipelines
RL + GenAI integration, AI safety
Integrated AI system (student-designed)
AI
Master AI through practical application
Practical Skills
Build real-world applications and projects
Problem Solving
Develop computational thinking and problem-solving abilities
Join this course and start your learning journey today.
