Deep Learning RNNaissance

Main talk

Jürgen Schmidhuber (Swiss AI Lab IDSIA) - Deep Learning RNNaissance

Machine learning and pattern recognition are currently being revolutionised by "Deep Learning"(DL) Neural Networks (NNs). This is of commercial interest (for example, Google spent over 400m on start-up "deepmind" co-founded by our student). My talk will summarise our work on DL since 1991. Our recurrent NNs (RNNs) were the first to win official international competitions in pattern recognition and machine learning; our team has won more such contests than any other research group or company. In particular, our RNNs represent the state of the art in connected handwriting recognition, and aspects of speech recognition. Our Deep Learners also were the first to win object detection and image segmentation contests. We also built the first artificial RNN-based agent that learns from scratch complex control based on high-dimensional vision. Time permitting, I'll also address curious/creative machines and theoretically optimal, universal, self-modifying artificial intelligences.

Lightning talk

Abe Stanway (Planet Labs) - Quick and Dirty Timeseries Techniques

Timeseries are everywhere, and some solid timeseries analysis chops are an essential part of the modern data scientist's toolbox. In this talk, I'll be going over a few established machine learning techniques for timeseries analysis, including anomaly detection, changepoint detection, and prediction. I'll also cover some less-established-but-still-promising ideas including causal inference through graphical models.

Main speaker bio

Since age 15 or so, Prof. Jürgen Schmidhuber's main scientific ambition has been to build an optimal scientist through self-improving AI, then retire. His AI team has won nine international competitions in machine learning & pattern recognition and seven independent best paper/best video awards, achieved the world's first superhuman visual classification results, and established the field of mathematically rigorous universal AI and optimal universal problem solvers. His formal theory of creativity & curiosity & fun explains art, science, music, and humor. He also generalized algorithmic information theory and the many-worlds theory of physics, and introduced the concept of Low-Complexity Art, the information age's extreme form of minimal art. Many famous companies are now using the techniques developed in his group at the Swiss AI Lab IDSIA (a Business Week Top 10 AI Lab already back in 1997) & USI & SUPSI. Since 2009 he is member of the European Academy of Sciences and Arts. He published over 300 peer-reviewed papers, and is recipient of the 2013 Helmholtz Award of the International Neural Networks Society. Home page: http://www.idsia.ch/~juergen/

Tentative Schedule:

6:30pm - 7:00pm -- socializing

7:00pm - 7:20pm -- lightning talk (Abe Stanway)

7:20pm - 8:20pm -- main talk (Jürgen Schmidhuber)

8:20pm - 8:00pm -- socializing

Special thanks to:

http://photos1.meetupstatic.com/photos/event/b/b/e/6/600_394728102.jpeg

• Upsight (http://www.upsight.com) for hosting + food/drinks

About Upsight:

Upsight’s mission is to transform the world’s data into valuable action. Mobile and social developers leverage Upsight’s comprehensive analytics and marketing platform to better understand user behavior, decide what it means, and act to impact their key business goals.

Formerly Kontagent+PlayHaven, Upsight now spans 25,400 apps, 700+ million monthly active users, 1.2+ billion devices, and 500+ billion in-app events each month.

Upsight is widely adopted by a variety of mobile-focused businesses including established brands such as Warner Brothers, Coffee Meets Bagel, and Turner, and game publishing powerhouses such as Glu, Electronic Arts, SEGA, and BigFish, as well as thousands of indie app developers.

Upsight is headquartered in San Francisco. For more information, please visit www.upsight.com (http://www.upsight.com/).