How deep learning allowed computers to see read more at here www.spinonews.com/index.php/item/1110-how-deep-learning-allowed-computers-to-see
One of the biggest challenges of the 21st century is to make computers more similar to the human brain. We want them to speak, understand and solve problems — and now we want them to see and recognize images. Computers can now train themselves on massive image databases to be able to identify key features, and without human intervention
Understanding machine learning is quite easy. The idea is to train algorithms on large databases to make them able to predict results from new data.
For a computer, a picture is millions of pixels — that’s a lot of data to process and too many inputs for an algorithm. Researchers had to find a shortcut. The first solution was to define intermediary characteristics.
According to the Techcrunch report, In the 2000s, Fei-Fei Li, director of Stanford’s AI Lab and Vision Lab, had a good intuition: How do children learn object names? How are they able to recognize a cat or a dress? Parents do not teach this by showing characteristics, but rather by naming the object/animal every time their child sees one. They train kids by visual examples. Why couldn’t we do the same for computers?
Imagine you want computers to recognize a cat. First of all, a human has to define all the main features of a cat: a round head, two sharp hears, a muzzle… Once the key features are defined, a well-trained neural network algorithm will, with a sufficient level of accuracy, analyze them and determine if the picture is a cat.
It is just the beginning for deep learning: We managed to make computers see as a three-year-old child, but, as Li said in a TED talk, “the real challenge is ahead: How can we help our computer to go from three to 13-year-old kid and far beyond?”
Read the full story here.
Comments
Post a Comment