
Sign up to save your podcasts
Or


Deep Learning - Reinforcement Learning Part 1
This video explains the concepts of sequential decision making and the multi-armed bandit problem.
Further Reading:
Deep Learning - Reinforcement Learning Part 2
This video explains the basics of reinforcement learning: The Markov Decision Process and how to compure the expected future return.
Further Reading:
Deep Learning - Reinforcement Learning Part 3
This video explains a fundamental learning technique in reinforcement learning: Policy Iteration.
Further Reading:
Deep Learning - Reinforcement Learning Part 4
This video discusses several other solution approaches to learning games such as Monte Carlo Techniques, Temporal Difference Learning, Q Learning, and learning universal function approximators for reinforcement learning using the policy gradient.
Further Reading:
Deep Learning - Reinforcement Learning Part 5
In the last video on reinforcement learning, we look into the deep reinforcement learning techniques. We start looking into how Deep Mind beat Atari Games and in particular breakout. Furthermore, we look into the technology behind AlphaGo and AlphaGoZero to play Go, Chess, and Shogi on world-class level.
Links
Video References:
Further Reading:
References
From the publisher's feed
(multilayer) perceptron, backpropagation, fully connected neural networks
loss functions and optimization strategies
convolutional neural networks (CNNs)
activation functions
regularization strategies
common practices for training and evaluating neural networks
visualization of networks and results
common architectures, such as LeNet, Alexnet, VGG, GoogleNet
recurrent neural networks (RNN, TBPTT, LSTM, GRU)
deep reinforcement learning
unsupervised learning (autoencoder, RBM, DBM, VAE)
generative adversarial networks (GANs)
weakly supervised learning
applications of deep learning (segmentation, object detection, speech recognition, ...)
The accompanying exercises will provide a deeper understanding of the workings and architecture of neural networks.

0 Listeners

0 Listeners

2 Listeners

0 Listeners

1 Listeners

0 Listeners

6 Listeners

0 Listeners

0 Listeners