Reinforcement Learning: Demystifying the hype
summary
Covers reinforcement learning (RL) and its fusion with deep learning (Deep Q-learning, AlphaGo Zero) to handle high-dimensional inputs from raw images and learn by trial-and-error using reward signals. Explains how to formulate practical problems as RL tasks and surveys real-world industry applications (robotics, automated trading, manufacturing, energy, dialog systems, recommendation engines), emphasizing when RL is appropriate and showing successful use cases.
AI-written summary · about
chapters (10)
- 0:00Start
- 20:16Markov Decision Process
- 21:39Gridworld
- 27:37Robotics
- 27:58Dialogue Systems
- 29:32Trading System
- 30:50Multi Armed Bandits
- 33:19MAB Formulation
- 36:17Recommendation Engines
- 37:41Closing thought
related talks
05mPractical insights into deep reinforcement learning
44mReinforcement learning: A gentle introduction
43mIntroduction to (deep) reinforcement learning
30mReinforcement Learning - ChatGPT, Playing Games & More
33mBreakthroughs and the Future of (Deep) Reinforcement Learning
01h 31mReinforcement Learning for Finance
26mAnkur Narang; Senior Vice President – Technology & Decision Sciences at Yatra at Cypher2017- Part1/3
38mReinforcement Learning - What it is and How it Works