When scientists test algorithms that sort or classify data, they often turn to a trusted tool called Normalized Mutual Information (or NMI) to measure how well an algorithm's output matches reality.

In the lower part of the figure, it can be seen that L2O leverages on a set of training problem instances from the target optimization problem class to gain knowledge. This knowledge can help identify ...

The original version of this story appeared in Quanta Magazine. For computer scientists, solving problems is a bit like mountaineering. First they must choose a problem to solve—akin to identifying a ...