Master Q-learning, policy gradients, and deep reinforcement learning to build autonomous agents that learn from their environment and make intelligent decisions.
Dive deep into reinforcement learning and create intelligent agents that learn optimal behaviors through trial and error. Master cutting-edge algorithms used in robotics, gaming AI, and autonomous systems.
Markov Decision Processes, value functions, Bellman equations, and dynamic programming
Q-learning, SARSA, expected SARSA, and function approximation
REINFORCE, actor-critic, advantage functions, and natural policy gradients
DQN, Double DQN, Dueling DQN, Rainbow, and experience replay
PPO, A3C, DDPG, TD3, SAC, and continuous control
Multi-agent reinforcement learning, imitation learning, and inverse RL
Design and implement a complete RL system for a real-world application
Research Scientist at DeepMind
PhD in Machine Learning from Stanford, 12+ years in RL research, co-author of 50+ papers, key contributor to AlphaGo and modern RL algorithms.
Train agents to master Atari games using deep Q-networks and policy gradients.
DQN + PPODevelop continuous control algorithms for robotic arm manipulation and locomotion.
DDPG + SACBuild an intelligent trading agent that learns optimal investment strategies from market data.
Multi-Agent"Dr. Kumar's course gave me the deep understanding I needed to transition into AI research. I'm now working on autonomous vehicles at Waymo!"
"The hands-on projects were incredibly challenging and rewarding. The trading agent I built got me noticed by quantitative trading firms!"