• Linear regression
• Logistic regression
• Fully-connected network
• Convolutional network
• Variational auto-encoder
• Deep Deterministic Policy Gradient (DDPG)

• Simple network achieving 88% testing accuracy on cifar10
• Transfer learning from tiny-imagenet to cifar10
• Train a network to approximate XOR and MAX function

TODO: Implement various deep-rl algorithms including

• Deep Q-Networks (DQN)
• DDPG (Deep Deterministic Policy Gradient)
• A3C (Asynchronous Advantage Actor-Critic)
• TRPO (Trust Region Policy Optimization)
• PPO (Proximal Policy Optimization)