The Bluestocking, Vol XXI: Playing Go, winning at Twitter and building a girl
helenlewis.substack.com
Happy Friday everyone! There will be no newsletter next week, for reasons that will be explained later *swirls cape*. Helen Showdown Deep learning requires two things: plenty of processing grunt and plenty of data to learn from. DeepMind trained its machine on a sample of 30m Go positions culled from online servers where amateurs and professionals gather to play. And by having AlphaGo play against another, slightly tweaked version of itself, more training data can be generated quickly.
The Bluestocking, Vol XXI: Playing Go, winning at Twitter and building a girl
The Bluestocking, Vol XXI: Playing Go…
The Bluestocking, Vol XXI: Playing Go, winning at Twitter and building a girl
Happy Friday everyone! There will be no newsletter next week, for reasons that will be explained later *swirls cape*. Helen Showdown Deep learning requires two things: plenty of processing grunt and plenty of data to learn from. DeepMind trained its machine on a sample of 30m Go positions culled from online servers where amateurs and professionals gather to play. And by having AlphaGo play against another, slightly tweaked version of itself, more training data can be generated quickly.