What was the Google Brain project, and how did it famously teach itself to recognize cats?
Google gave a neural network 10 million YouTube thumbnails. It became obsessed with cats - unsupervised.
The answer
The Google Brain project, founded in 2011 by Andrew Ng and Jeff Dean as part of Google's X Lab, aimed to explore the potential of large-scale deep learning using Google's massive computing infrastructure. In June 2012, the team made headlines when their network of 16,000 CPU cores, trained on 10 million random YouTube thumbnails using unsupervised learning, spontaneously developed neurons that responded strongly to human faces and cats - without ever being told what a face or cat was. The project demonstrated that scale could enable emergence in neural networks. Google Brain eventually grew into one of the world's premier AI research groups and later merged with DeepMind in April 2023.
Source: arXiv (Le et al., 2012)
- google brain
- unsupervised learning
- large-scale computing
More from AI & Machine Learning
- What radical question did Alan Turing ask in 1950 that launched the entire field of artificial intelligence?
- Can a machine fool you into thinking it's human - and does that mean it's truly intelligent?
- What happened at a small Dartmouth workshop in 1956 that gave artificial intelligence its name?
- How did a 1958 invention called the Perceptron convince people that machines could learn like brains?
- Could a 1966 chatbot really make people believe they were talking to a therapist?
- What was the first robot that could 'reason' about its own actions?