DeepMind
DeepMind is a division of Alphabet, Inc. responsible for developing general-purpose artificial intelligence (AGI) technology. That technology is also known as Google DeepMind.
DeepMind uses raw pixel data as input and learns from experience. The AI uses deep learning on a convolutional neural network, with a model-free reinforcement learning technique called Q-learning.
While the idea of general purpose AI is controversial, Google set out to establish and improve their AI property on a wide variety of grounds. DeepMind technology has been challenged to learn games on its own. For example, when it was tasked to beat the library of Atari games, it learned to understand the games without changing the code. After a time, the AI could play the games better and with more efficiency than humans.
In pushing the boundaries of AI, DeepMind tasked itself with defeating the board game Go. Go is a computational challenge for AI, largely because of the complexity of choosing among the immense number of possible moves in the game. DeepMind developed a special project called AlphaGo, a computer program designed to play the board game. After numerous versions of supervised learning AI models, AlphaGo bested the No. 1 player worldwide in 2017. In the same year, DeepMind released a new version named AlphaGo Zero, using unsupervised re-enforcement learning, which consistently bested previous versions of the program.
Outside of playing games, DeepMind was used to improve power efficiency in the already optimized data centers at Google. DeepMind was able to improve on the efforts of previous specialists by 15%, making a 40% reduction in cooling costs. It’s also been used in developing Google Assistant and helps create personalized app recommendations in Google Play.
DeepMind was created in 2010 in London by Demis Hassabis, Shane Legg