Google's DeepMind team have created a human like memory for their AI

Memory fills the gaps in information that even todays best AI algorithms can't Google's DeepMind artificial intelligence lab does more than just develop computer programs capable of beating the world's best human players in the ancient game of Go. The DeepMind unit has also been working on the next generation of deep learning software that combines the ability to recognize data patterns with the memory required to decipher more complex relationships within the data. Now, it's at this point where when we say "memory" you might think that we're referring to the type of RAM or some other form of hardware computer memory you stick into servers, but no. After all, say for example we asked you to predict an outcome then it's likely that not only will you draw on the information you have on hand at the time but you'll also – consciously or sub-consciously – draw on past memories that, when accessed and interpreted correctly, will help you create a better, more informed answer – or opinion. It's this "human" type memory – this "external" memory as researchers call it – that we're increasingly seeing being used to augment today's most advanced deep learning AI's – such as DeepMind. Deep learning is the latest buzz word for artificial intelligence algorithms called neural networks that can learn over time by filtering huge amounts of relevant data through many "deep" layers.