Develop AI systems to efficiently learn new skills and solve open-ended problems, rather than depend exclusively on AI systems trained with extensive datasets
Discover how this asset is connected to other assets on the platform
This one's purely a recoloring exercise. But the rules to recolor are not so simple and require some reasoning to figure out.
The rule is: for forms with 3 or more pieces, keep the color. Otherwise, transform to green.
The rule set is so simple, but it took a little time to figure out because it's not a direct or linear transformation. Some structures changed color, while others did not.
Once we figure out reasoning, I think this is something that computers may actually do better at because they won't be expecting some direct mapping like we often do as humans.
Discover assets like this one.
No physics-based approach is going to help us here. Perhaps understanding what we consider a structure (which does change between challenges) may be somewhat physics-inspired. Understanding what is a structure requires us to test different possible structure rules on the input seeing where we find the structure maintained in the output (even if some characteristic about it has changed, like color).
It seems reasoning is the application of many mini algorithms; and deciding what the algorithms are and when to apply them is part of the task.
In this one, my brain wrote a structure finding algorithm because that was the first part of understanding what might get recolored and how.