This seminar offers an exciting introduction to Deep Reinforcement Learning (RL), a field at the intersection of machine learning and decision-making. In RL, an agent learns to act optimally in an environment by trial and error, receiving rewards as feedback. Deep RL supercharges this process by using neural networks to approximate value functions and policies, enabling breakthroughs in high-dimensional tasks like robotic manipulation, autonomous driving, and mastering complex video games beyond human-level performance. Throughout the seminar, we will dive into key RL algorithms and address fundamental challenges such as sample efficiency, exploration, and credit assignment---critical hurdles in scaling RL to real-world problems. We will also discuss recent advances, including the role of large-scale models, offline RL, and reinforcement learning with human feedback (RLHF). A solid understanding of probability and linear algebra is expected, and familiarity with Python is highly recommended.