Learning Abstract Classes using Deep Learning
This addresses the problem of AI systems learning abstract concepts, which is crucial for advancing machine perception, but it appears incremental as it applies an existing method to a new task.
The paper tested a convolutional neural network (GoogLeNet) on differentiating abstract classes like horizontal and vertical orientation, finding that it could transfer learned classes to unseen objects, though performance details are not specified.
Humans are generally good at learning abstract concepts about objects and scenes (e.g.\ spatial orientation, relative sizes, etc.). Over the last years convolutional neural networks have achieved almost human performance in recognizing concrete classes (i.e.\ specific object categories). This paper tests the performance of a current CNN (GoogLeNet) on the task of differentiating between abstract classes which are trivially differentiable for humans. We trained and tested the CNN on the two abstract classes of horizontal and vertical orientation and determined how well the network is able to transfer the learned classes to other, previously unseen objects.