ADLxMLDS2017 The projects we did in lecture: Application of deep learning in 2017 fall. I got A+ in this class Lecture link Projects Sequence_Labeling Using TIMIT mfcc data predicting phone sequence Join the kaggle competition and I got rank 36th/191 Kaggle link Video_Captioning Implement s2vt model to predict the behavior in video Using attention mechanism improve performance Game_Playing Using reinforcement learning play atari game Deep Q learning model for breakout-v1 Gradient policy & A2C model for pong Comics_Generation text to Comics face image Using DCGAN model and conditional GAN I got rank 5th/191 Implementing cycle gan for style transfer is our bouns Final Project lf2gym We use our novel model (A3C+) training on 2.5D game (little fighter 2) We build a fine environment for our agent Our agent well defeats original AI