The Need for Speed: Effects of Human Derived Time Constraints on Performance and Strategy in Machine Models of Tetris
One of the hallmarks of expert performance in complex, dynamic tasks is the ability to select and perform the appropriate action within a constantly shifting environment, often under tight time constraints. In an example task, the video game Tetris, expert players select placement positions for the active zoid and navigate them into place in increasingly short time spans. Machine models of the same task are capable of producing human-like performance patterns, but either ignore or only roughly approximate the time constraints that seem to be an integral part of human behavior. Using a set of scaled time parameters derived from a large set of human subjects, we trained and tested an existing machine Tetris model and observed the resultant changes in performance and behavior.
hi Catherine! Awesome talk and model! I was wondering whether you have considered looking at the effect of suboptimal guesses of how long it will take to make a move, or slowness in making the move, both of which I would consider to make a difference in real humans. How would that affect the model's performance?
Thanks for the presentation! I found this project very interesting. I have a few follow-up questions for you: 1) Table 6 suggests that models trained under "expert" level time pressure were able to get higher proportions of multi-line clears than models trained under "extreme expert" level time pressure. Similarly, models trained under "interme...
Cite this as: