Explore deep Q-networks in this comprehensive 41-minute lecture by Pascal Poupart. Delve into key concepts including approximation, optimization, Q-learning, gradient Q-learning, and experience replay. Learn about divergence issues and their solutions, gradient descent techniques, and the implementation of target networks. Discover the architecture and algorithm behind deep Q-networks, and examine their performance results in various applications.
Overview
Syllabus
Intro
Overview
Approximation
Optimization
QLearning
QLearning Recap
Gradient QLearning
Divergence
Experience Replay
Gradient Descent
Target Network
Deep Queue Network
Deep Queue Network Algorithm
Deep Queue Network Architecture
Deep Queue Network Results
Taught by
Pascal Poupart
Related Courses
-
Deep Reinforcement Learning in Python
-
Deep Neural Network for Beginners Using Python
-
Deep Learning - IIT Ropar
-
Deep Learning Interview Prep Course
-
Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization
5.0 -
How Neural Networks Learn: Exploring Architecture, Gradient Descent, and Backpropagation
