The following Python script is used to produce the movements of a game AI character in a videogame.

#!/usr/bin/env python3
import time,random
for i in range(100000):

In contrast to a human player, the script doesn't try to maximize it's reward, instead it's repeating the same actions over and over again. The script doesn't produce a sense and the hypothesis is, that the AI didn't have understand what the game objective is. Are random generators playing games different from humans, in a way that the script is flipping the social roles and avoids to become a participant of a game?


Random Decisions

Random decisions are sort of counter to the idea of AI, here understood as rational decision-making based on analysis of data, except in a limited number of cases:

  1. Randomness is utilized search (MCTS) and the the results are analyzed to determine the choice with the highest probability of returning the greatest utility

  2. An AI analyzes a game's equilibria and determines that a random choice is the optimal strategy (matching pennies).

The use of Random Number Generation in the example is no different than a dice roll--it's AI to the extent that it's a automated decision, but the strength of the decision making agent is unlikely to be high except where random choice is the most optimal strategy.

Goals & Understanding the Objective

An AI does not necessarily need to understand the objective of a given game, provided that local goals are sufficient to make decisions, and potentially approach a global optimum. (For instance, a Chess engine not based on checkmate, but where the objective simply removing as much of the opponent's materiel as possible, which could result in a victory against a weak opponent.)

Social Role/Human-like

Any game AI engaged in play can be considered to be in a social role, since games involving more than one player are communal activities. (This has nothing to do with random choices--humans themselves may choose to make random choices.) The AI is participating in a game, regardless of the rationale, or lack thereof, of a given decision.

  • $\begingroup$ quote “An AI does not necessarily need to understand the objective of a given game” Does this mean, that an AI can translate a letter into perfect English, without understanding English at all? $\endgroup$ – Manuel Rodriguez Sep 17 at 14:28
  • $\begingroup$ @ManuelRodriguez A strong translation might require the context of the entire letter, where phrase-by-phrase translation out-of-context might increase chance of errors. But it could be similar in terms of a global goal vs. local goals. $\endgroup$ – DukeZhou Sep 17 at 18:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.