About this episode
<p>Video games have become a favourite tool for AI researchers to test the abilities of their systems. In this episode, Hannah sits down to play StarCraft II - a challenging video game that requires players to control the onscreen action with as many as 800 clicks a minute. She is guided by Oriol Vinyals, an ex-professional StarCraft player and research scientist at DeepMind, who explains how the program AlphaStar learnt to play the game and beat a top professional player. Elsewhere, she explores systems that are learning to cooperate in a digital version of the playground favourite ‘Capture the Flag’.</p><p>If you have a question or feedback on the series, message us on Twitter (<a href="https://twitter.com/deepmindai?lang=en">@DeepMind</a> using the hashtag #DMpodcast) or emailing us at <a href="mailto:podcast@deepmind.com">podcast@deepmind.com</a>.</p><p><strong>Further reading</strong></p><ul><li><a href="https://www.economist.com/science-and-technology/2017/05/13/why-ai-researchers-like-video-games">The Economist: Why AI researchers like video games</a></li><li>DeepMind blogs: <a href="https://deepmind.com/blog/article/capture-the-flag-science">Capture the Flag</a> and <a href="https://deepmind.com/blog/article/alphastar-mastering-real-time-strategy-game-starcraft-ii">Alphastar</a></li><li><a href="https://www.youtube.com/watch?v=zgIFoepzhIo&feature=youtu.be">Professional StarCraft II player MaNa gives his impressions of AlphaStar and DeepMind</a></li><li><a href="https://openai.com/five/">Open AI’s work on Dota 2 </a></li><li><a href="https://www.nytimes.com/2019/05/30/science/deep-mind-artificial-intelligence.html">The New York Times: DeepMind can now beat us at multiplayer games, too</a></li><li><a href="https://royalsociety.org/topics-policy/projects/machine-learning/?utm_source=royalsociety.org&utm_medium=redirect&utm_campaign=machine-learning">Royal Society: Machine Learning resources</a></li><li><a href="https://www.youtube.com/watch?v=UuhECwm31dM">DeepMind: The Inside Story of AlphaStar </a></li><li><a href="http://karpathy.github.io/2016/05/31/rl/">Andrej Karpathy: Deep Reinforcement Learning: Pong from Pixels</a></li></ul><p><strong>Interviewees:</strong> Research scientists Max Jaderberg and Raia Hadsell; Lead researchers David Silver and Oriol Vinyals, and Director of Research Koray Kavukcuoglu.</p><p><strong>Credits:</strong><br />Presenter: Hannah Fry<br />Editor: David Prest<br />Senior Producer: Louisa Field<br />Producers: Amy Racs, Dan Hardoon<br />Binaural Sound: Lucinda Mason-Brown<br />Music composition: Eleni Shaw (with help from Sander Dieleman and WaveNet)<br />Commissioned by DeepMind</p> <p><p>Please leave us a review on Spotify or Apple Podcasts if you enjoyed this episode. We always want to hear from our audience whether that's in the form of feedback, new idea or a guest recommendation! </p><p> </p></p><br/> <p>Hosted by Simplecast, an AdsWizz company. See <a href="https://pcm.adswizz.com">pcm.adswizz.com</a> for information about our collection and use of personal data for advertising.</p>