Motoki WuThe Bidirectional Language ModelEasy trick to include both left and right contextsFeb 8, 20185Feb 8, 20185
Motoki WuinTowards Data ScienceAdding Uncertainty to Deep LearningHow to construct prediction intervals for deep learning models using Edward and TensorFlowMar 10, 20172Mar 10, 20172
Motoki WuVisualizations of Recurrent Neural NetworksOpening up the black boxOct 27, 20161Oct 27, 20161
Motoki WuinTowards Data ScienceMachine Learning, Fast and SlowMemorization and generalization of modelsSep 12, 2016Sep 12, 2016
Motoki WuUnreasonable Effectiveness of CountingApproximating deep learning with simple counts*Jul 18, 20161Jul 18, 20161
Motoki WuMoar Data, Moar ErrorFriends don't let friends get drunk unless it's data.Sep 8, 2015Sep 8, 2015