
Sign up to save your podcasts
Or


Deep Learning - Architectures Part 1
This video discusses the first early architectures developed in deep learning from LeNet to GooLeNet.
Further Reading:
Deep Learning - Architectures Part 2
This video discusses the success of deeper models including Inception V2 and V3. One key technology that is introduced in V3 is label smoothing regularization.
Further Reading:
Deep Learning - Architectures Part 3
This video discusses the ideas of residual connections in deep networks that allow going from 20 to more than 1000 layers.
Further Reading:
Deep Learning - Architectures Part 4
This video demonstrates the many uses of residual connections in deep networks from Inception-ResNet to Densenet.
Further Reading:
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