Unsupervised Learning | DeepMindgoogleplusDMlinkedinredditsearchsearchtwitteryoutubegoogleplusfullscreenfullscreen_mobilefullscreenfullscreen_mobilefullscreenfullscreen_mobiletwitterfacebookgoogle plu

Unsupervised Learning | DeepMindgoogleplusDMlinkedinredditsearchsearchtwitteryoutubegoogleplusfullscreenfullscreen_mobilefullscreenfullscreen_mobilefullscreenfullscreen_mobiletwitterfacebookgoogle plu

Unsupervised learning is a paradigm designed to create autonomous intelligence by rewarding computer programs for learning about the data they observe without a particular task in mind--in other words, the program learns for the sake of learning. We believe unsupervised learning will be foundational to building artificial general intelligence.

49 mentions: @DeepMindAI@amritamahale@JMateosGarcia@__tmats__@stevenk11011@RuthBarnett@sjoerdapp@AiLeadership
Date: 2019/04/10 14:18

Referring Tweets

@DeepMindAI Our new blog post overviews unsupervised learning, a paradigm for creating artificial intelligence that learns about data without a particular task in mind. Read more about how we might teach computers to learn for the sake of learning: t.co/BmZU5dsih2
@JMateosGarcia If you ever need to efficiently explain the difference between supervised, reinforcement and unsupervised learning, this @DeepMindAI blog has your back. t.co/OvW516ZBJG t.co/XHKrQkeHpm
@amritamahale “Unsupervised learning is a paradigm designed to create autonomous intelligence by rewarding agents for learning about the data they observe without a particular task in mind. the agent learns for the sake of learning.” Fascinating piece. t.co/5oe6KcHLXF
@RuthBarnett If you've ever found yourself baffled / curious about the language used in AI research, this explainer (by some colleagues) is a great place to start. t.co/NLGzC8kZcI
@AiLeadership New blog post overviews unsupervised learning, a paradigm for creating artificial intelligence that learns about data without a particular task in mind. Read more about how we might teach computers to learn for the sake of learning: t.co/iDRuyTFTRr … via @DeepMindAI
@im_prashant_rai Unsupervised learning would be a major breakthrough for artificial general intelligence. It's induced by finding patterns from available data without having a particular task(targets) to do. #unsupervised_learning, #deepmind_blog t.co/uKUrT1MsjA
@luckflow Our new blog post overviews unsupervised learning, a paradigm for creating artificial intelligence that learns about data without a particular task in mind. Read more about how we might teach computers to learn for the sake of learning: t.co/li49R8hXCr
@__tmats__ DeepMindのブログ,ついにDeepMindからも"world model"という言葉が登場してる Unsupervised Learning t.co/4qmWEBp2NN

Related Entries

Read more Interview with Kaggle Grandmaster, Lead Data Scientist at Dbrain: Artur Kunzin
0 users, 4 mentions 2019/02/23 05:26
Read more Towards Robust and Verified AI: Specification Testing, Robust Training, and Formal Verification | De...
3 users, 46 mentions 2019/03/28 18:47
Read more DeepMind and Waymo: how evolutionary selection can train more capable self-driving cars | DeepMindgo...
0 users, 36 mentions 2019/07/25 14:17
Read more CCMatrix: A billion-scale bitext data set for training translation models
2 users, 18 mentions 2020/02/07 15:50
Read more RE•WORK | AI Applications Virtual Summit
0 users, 6 mentions 2020/08/13 18:53