3 ms·
the OP is referring to the use of a GAN, which is a different type of setup. GAN's were not used for alphaGo, as the article points out Deepmind uses reinforce
by Barrin92 9y ago
the OP is referring to the use of a GAN, which is a different type of setup.
GAN's were not used for alphaGo, as the article points out Deepmind uses reinforcement learning and MCTS.
- nschucher 9y agoThey're most likely referring to adversarial attacks where degenerate inputs are constructed that could cause AlphaGo Zero to perform sub-optimally or catastrophically fail (see OpenAI Research [0]). This is distinct from generative adversarial networks (GANs) or adversarial self-play (which I guess AlphaGo Zero is an example of). [0] https://blog.openai.com/adversarial-example-research/ https://blog.openai.com/adversarial-example-research/