- Duration: 3 days
- Skill-level: Foundation-level Deep Reinforcement Learning skills for Intermediate skilled team members. This is not a basic class.
- Targeted Audience: This course is geared for Python developers, analysts or others who wants a powerful machine learning approach, using examples, illustrations, exercises, and crystal-clear teaching. You’ll love the perfectly paced teaching and the clever, engaging writing style as you dig into this awesome exploration of reinforcement learning fundamentals, effective deep learning techniques, and practical applications in this emerging field.
- Hands-on Learning: This course is approximately 50% hands-on lab to 50% lecture ratio, combining engaging lecture, demos, group activities and discussions with machine-based student labs and exercises. Student machines are required.
- Delivery Format: This course is available for onsite private classroom presentation, or remote instructor led delivery, or CBT/WBT (by request).
- Customizable: This course may be tailored to target your specific training skills objectives, tools of choice and learning goals.
Deep Reinforcement Learning is a beautifully balanced approach to teaching, offering numerous large and small examples, annotated diagrams and code, engaging exercises, and skillfully crafted writing. You’ll explore, discover, and learn as you lock in the ins and outs of reinforcement learning, neural networks, and AI agents. You will go from small grid world environments and some of the foundational algorithms to some of the most challenging environments out there today and cutting-edge techniques to solve these environments.
Working in a hands-on learning environment, led by our Deep Reinforcement expert instructor, students will learn about and explore:
- You’ll explore, discover, and learn as you lock in the ins and outs of reinforcement learning, neural networks, and AI agents.
- You will go from small grid world environments and some of the foundational algorithms to some of the most challenging environments out there today and cutting-edge techniques to solve these environments.
Topics Covered: This is a high-level list of topics covered in this course. Please see the detailed Agenda below
- Foundational reinforcement learning concepts and methods
- The most popular deep reinforcement learning agents solving high-dimensional environments
- Cutting-edge agents that emulate human-like behavior and techniques for artificial general intelligence
Audience & Pre-Requisites
This course is geared for attendees with Python skills who wish to build machine learning systems that explore and learn based on the responses of the environment.
Pre-Requisites: Students should have
- developers with some understanding of deep learning algorithms.
- Experience with reinforcement learning is not required.
- Perfect for readers of Deep Learning in Python or Deep Learning.
- Good foundational mathematics or logic skills
- Basic Linux skills, including familiarity with command-line options such as ls, cd, cp, and su
Course Agenda / Topics
- Introduction to deep reinforcement learning
- What is deep reinforcement learning?
- The past, present, and future of deep reinforcement learning
- The suitability of deep reinforcement learning
- Setting clear two-way expectations
- Mathematical foundations of reinforcement earning
- Components of reinforcement learning
- MDPs: The engine of the environment
- Balancing immediate and long-term goals
- The objective of a decision-making agent
- Planning optimal sequences of actions
- Balancing the gathering and utilization of information
- The challenge of interpreting evaluative feedback
- Strategic exploration
- Evaluating agents’ behaviors
- Learning to estimate the value of policies
- Learning to estimate from multiple steps
- Improving agents’ behaviors
- The anatomy of reinforcement learning agents
- Learning to improve policies of behavior
- Decoupling behavior from learning
- Achieving goals more effectively and efficiently
- Learning to improve policies using robust targets
- Agents that interact, learn and plan
- Introduction to value-based deep reinforcement learning
- The kind of feedback deep reinforcement learning agents use
- Introduction to function approximation for reinforcement learning
- NFQ: The first attempt to value-based deep reinforcement learning
- More stable value-based methods
- DQN: Making reinforcement learning more like supervised learning
- Using experience replay
- Double DQN: Mitigating the overestimation of action-value functions
- Sample-efficient value-based methods
- Dueling DDQN: A reinforcement-learning-aware neural network architecture
- network brings to the table?
- PER: Prioritizing the replay of meaningful experiences
- Policy-gradient and actor-critic methods
- REINFORCE: Outcome-based policy learning
- VPG: Learning a value function
- A3C: Parallel policy updates
- GAE: Robust advantage estimation
- A2C: Synchronous policy updates
- Advanced actor-critic methods
- DDPG: Approximating a deterministic policy
- TD3: State-of-the-art improvements over DDPG
- SAC: Maximizing the expected return and entropy
- Towards artificial general intelligence
- What was covered, and what notably wasn’t?
- More advanced concepts towards AGI
- Fairness, and Ethical Standards
- What happens next?