Check out the PR comments on one of his projects on GH.
> 'Missing Implementation for "Deny, Defend, Depose"'
"Could you clarify:
What machine learning model is best suited for the "Deny" phase? (I'm guessing a GAN for plausible deniability?)
Is "Defend" using a reinforcement learning approach with adversarial training?
For "Depose," are we looking at a classification problem or full-blown unsupervised chaos?
A README section explaining how to integrate this strategy into Halite III bots would be invaluable to the community. Until then, I feel like I'm shooting blanks here."
> 'Missing Implementation for "Deny, Defend, Depose"'
"Could you clarify:
What machine learning model is best suited for the "Deny" phase? (I'm guessing a GAN for plausible deniability?) Is "Defend" using a reinforcement learning approach with adversarial training? For "Depose," are we looking at a classification problem or full-blown unsupervised chaos?
A README section explaining how to integrate this strategy into Halite III bots would be invaluable to the community. Until then, I feel like I'm shooting blanks here."
https://github.com/lnmangione/Halite-III/issues