You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In terms of game theory, if one of the participants chooses the strategy to make fully random decisions then the outcome of multiple games would be approximately the same amount of wins and losses on each side. And this outcome not depending on the decisions of the other party which makes the game boring.
If the bot decreases randomness in its strategy like doing Math.floor(Math.random() * Math.random() * 3), so the first move would be chosen more frequently then the user can guess it and play against this decision. It would be more interesting, but the bot would have a big risk of losing its coins.
A stronger strategy for the bot would be to assume that the opponent doesn't generate a good random, so to predict the next move by calculating the most frequent move based on the last 10 moves of the opponent, and to play against that move.
The bot can switch strategy during the game, keeping its balance at an acceptable level. If implementing something like this consider making the bot's source code private to avoid building the ideal user strategy based on it.
The text was updated successfully, but these errors were encountered:
state-channel-demo/server/src/services/contract/contract.service.ts
Lines 33 to 45 in 5bad5a3
In terms of game theory, if one of the participants chooses the strategy to make fully random decisions then the outcome of multiple games would be approximately the same amount of wins and losses on each side. And this outcome not depending on the decisions of the other party which makes the game boring.
If the bot decreases randomness in its strategy like doing
Math.floor(Math.random() * Math.random() * 3)
, so the first move would be chosen more frequently then the user can guess it and play against this decision. It would be more interesting, but the bot would have a big risk of losing its coins.A stronger strategy for the bot would be to assume that the opponent doesn't generate a good random, so to predict the next move by calculating the most frequent move based on the last 10 moves of the opponent, and to play against that move.
The bot can switch strategy during the game, keeping its balance at an acceptable level. If implementing something like this consider making the bot's source code private to avoid building the ideal user strategy based on it.
The text was updated successfully, but these errors were encountered: