It’s free, every week, in your inbox.
However, comparing neural networks to human perception remains a challenge.
The complex workings of deep learning systems also compound the problem.
Deep neural networkswork in very complicated waysthat often confound their own creators.
How do neural networks perceive contours?
The first test involves contour detection.
Can you tell which one of the above images contains a closed shape?
For humans, a closed contour flanked by many open contours perceptually stands out.
They usedtransfer learningto finetune the AI model on 14,000 images of closed and open contours.
They then tested the AI on various examples that resembled the training data and gradually shifted in other directions.
The initial findings showed that a well-trained neural internet seems to grasp the idea of a closed contour.
The model also seemed to struggle with detecting shapes when they became larger than a certain size.
The image on the right has been modified with adversarial perturbations, noise that is imperceptible by humans.
To the human eye, both images are identical.
But to a neural connection, they are different images.
Can machine learning reason about images?
The second experiment tested the abilities of deep learning algorithms in abstract visual reasoning.
The tests include same-different tasks (e.g., are two shapes in a picture identical?)
and spatial tasks (e.g., is the smaller shape in the center of the larger shape?).
A human observer would easily solve these problems.
The SVRT challenge requires the participating AI to solve same-different and spatial tasks.
The results show that a pretrained model finetuned on 28,000 samples performs well both on same-different and spatial tasks.
(Previous experiments trained a very small neural online grid on a million images.)
The researchers note that the human visual system is naturally pre-trained on large amounts of abstract visual reasoning tasks.
Consider the following image.
Can you tell what it is without scrolling further down?
Below is the zoomed-out view of the same image.
Theres no question that its a cat.
Deep learning systems also operate on features, but they work in subtler ways.
Previous experiments show a large difference between the image recognition gap in humans and deep neural networks.
These patches favor the human vision system.
And they draw conclusions that can provide directions for future AI research.
you might read the original articlehere.
