We have been hearing news of the artificial intelligence system outperforming humans in many tasks, some of the noticeable victories by AI are AlphaGo, DOTA2, StarCraft II etc. Reinforcement learning in a domain in machine learning requires the agent in the potentially complex environment to learn and achieve a goal under uncertainty. Reinforcement learning is applied to various fields like robotics, pattern recognition, personalized medical treatment, drug discovery, speech recognition and many more.
This workshop will learn about reinforcement learning and deep reinforcement learning, building games, and other essential algorithms in reinforcement learning. If you are a beginner-level data scientist or interested in learning advanced ML/AI topics, in that case, this course will provide you with an in-depth understanding of reinforcement learning theory and programming techniques.
FORMAT: Pre-recorded videos (More than 5 hours of content) & Colab notebooks
Pricing: $19.99
Note: the presentation used in the workshop will not be shared with participants.
Get the workshop videos at $19.99
After payment, you will receive an email with a download link to the whole workshop (Videos & notebook links).
ADaSci Members receive a 30% discount.
The workshop is free for CDS Charterholders.
Content
Introduction to Reinforcement Learning
- Basics of Learning paradigm
- Ingredients of Reinforcement Learning
- Agent
- Actions
- Environment
- Policy
- Markov Decision Processes and Bellman Equations
- Dynamic Programming
- Monte Carlo Methods
- Temporal Difference Learning
- Delayed Q-learning vs Double Q-learning vs Q-Learning
- N Step Algorithms
- Transfer and Multi-task Reinforcement Learning
- Imitation Learning and Inverse Reinforcement Learning
- Bayesian Reinforcement Learning
- Reinforcement Learning Applications
Introduction to Deep Reinforcement Learning
- What Is Deep Q-Learning?
- The Action-Value Function
- Predicting Rewards with the State-Value Function
- Predicting Rewards with the Action-Value Function
- Deep Reinforcement Learning Applications
Hands-on Reinforcement Learning Frameworks:
- OpenAI Gym
- Keras-RL
- Nervana Systems Coach
- Garage
- Surreal
- Tensorforce
- Google Dopamine
Introduction to Policy-Based methods
- Introduction to Value-Based and Model-Based Reinforcement Learning
- Introduction to Actor-Critic Model
- Policy Gradients
- Proximal Policy Gradients
- Deep Deterministic Policy Gradients
- Policy optimization
Implement Deep Q-network with PyTorch to play a game
- Environment set up
- Create and train DQN to play game
- Watch trained DQN play game
Hands-on Open Source toolkits for Deep Reinforcement Learning
- Facebook’s ReAgent
- RLCard
- RL-Medical for medical images
- RLlib
- OpenAI Baselines
Recent Advances and Applications
- Multi-Agent Reinforcement Learning
- Ethics in Reinforcement Learning
- Applying Reinforcement Learning for real-world problems
Key Takeaways:
– Hands-on experience on some of the most used frameworks on Reinforcement Learning
– Structure and create a reinforcement learning problem
– Learn about the latest advancement in Reinforcement Learning
– Building games using reinforcement learning
– Certificate on Build Your Reinforcement Learning Agent from Scratch
– Open problems in Reinforcement Learning
Prerequisites:
– Basic to moderate level python
– Basic of Pandas, Numpy, Scikit-learn and other Python packages
– Basic of Deep learning, CNN, RNN etc.
– Familiarity with Google Colab and GPU environment
Required Tools:
Any editor to run the python programs (preferably Google Colab Notebooks)
If working on an editor Pandas, Numpy, scikit-learn, TensorFlow, Pytorch and Keras must be installed.
High-speed internet connection
