View my Deep Learning Overview : [Google Slides]
Deep Learning Research Projects: [Google Slides]
Beware, these things get out of date very quick. This presentation is from Oct 2016.
The outline of the talk:
- Toy Neural Network
- Loss Function
- Stochastic Gradient Descent
- Forward-pass (Neural Function Evaluation)
- Backward-pass (Gradient of Neural function wrt to params)
- Recent Progress
- Convolutional Layer and its computations
- Pooling Layer and its computations
- Note on Non-linearity (ReLU, PReLU, Sigmoid, Tanh etc.)
- Param initialization and recent progress
- Note of Caffe and other alternatives
- Commonly used Network Architectures
- Recent Projects
- My Work – MapNets
- Deep Visualization
- R-CNN
- Generative Adversarial Networks
- Bed Room image generation
- Deep Art
- Images which fool the imagenet models
- Recurrent Nets
- Reinforcement Learning
- References
- Deep Learning Book
- Fei Fei Li’s CS251n @ Stanford