d294: NIPS DeepMind Papers

NIPS DeepMind Papershttps://deepmind.com/blog/deepmind-papers-nips-part-1/

  • Interaction Networks for Learning about Objects, Relations and Physics (Paper)
  • Strategic Attentive Writer for Learning Macro-Actions (Paper) (Video)
  • Matching Networks for One Shot Learning (Paper)
  • Safe and efficient off-policy reinforcement learning (Paper)
  • Blazing the trails before beating the path: Sample efficient Monte-Carlo planning (Paper)
  • Deep Exploration via Bootstrapped DQN (Paper) (Video)