quantuminformation/youtube-space-invaders

Issues

• Add pre-trained agent models

  #12 opened 6 years ago by generic-github-user
  6

• Add button to reset game

  #13 opened 5 years ago by generic-github-user
  2

• Add reward/loss over time graphs for actor and critic networks

  #11 opened 6 years ago by generic-github-user
  2

• Add speed control in agent interpreter panel

  #8 opened 6 years ago by generic-github-user
  1

• End game when ships reach bases

  #9 opened 6 years ago by generic-github-user
  1

• Use maxPooling2d instead of avgPooling2d

  #19 opened 6 years ago by generic-github-user
  0

• Add actor network training system

  #18 opened 6 years ago by generic-github-user
  0

• Add option to disable TensorFlow.js reinforcement learning agent (bot)

  #6 opened 6 years ago by generic-github-user
  0

• Add learning rate control

  #15 opened 6 years ago by generic-github-user
  0

• When in bot mode, also reset game when player wins

  #7 opened 6 years ago by generic-github-user
  0

• Add controllable limit to size of experience replay buffer

  #10 opened 6 years ago by generic-github-user
  1

• Add testing data for actor and critic networks

  #14 opened 6 years ago by generic-github-user
  0

• Add random noise to agent action outputs to allow exploration of possible action space

  #16 opened 6 years ago by generic-github-user
  0

• Add visualization of convolutional layer outputs

  #17 opened 6 years ago by generic-github-user
  0

• Agent API for tensorflow.js agent

  #4 opened 6 years ago by quantuminformation
  1

• Mobile-friendly

  #1 opened 7 years ago by quantuminformation
  0

• Embed on websites

  #2 opened 7 years ago by quantuminformation
  6