심화1팀

GAN

$$ \argmax_D $$

VAEs vs. GAN

Reinforcement Learning

Reward

$$ R_t = \sum_{i=t}^\infty \gamma^i r_t $$

DQN/Policy Gradient

$$ \mathcal L = -\log P(a_t | s_t) R_t $$

자연어기초팀

PyTorch Rhythm

Data Definition 단계

Text Representation

Bag of Words (BoW)

희소행렬(sparse matrix)로 나타내짐