In this talk, Neel will try to explore some of the concepts and ideas in deep learning that he finds most interesting, and which he thinks are most appealing to mathematical tastes. Machine Learning is a very young field, so he'll also explore some of the most fascinating areas of active research. Deep learning and neural networks are often presented as a mysterious black box, that just works by magic. But there has been a lot of recent conceptual and empirical progress in understanding what actually goes on inside neural networks, and Neel will especially focus on this.

This talk will try to be high-level and focus on the intuitions and conceptual understanding, rather than the fiddly details, so it will be fairly non-technical.

A basic understanding of the ideas behind machine learning is preferable (e.g. see the first half-hour of his previous talk on    • An Intuitive Introduction to Machine ...  ).

Visit Neel's personal webpage https://www.neelnanda.io/ml-talk for the slides and further reading!

And if you wish, any feedback to the speaker through https://forms.gle/wRpdL28eu6Qnn62u5 will be highly appreciated!

--------------------------------------------------------------------------------
Subscribe:    / @thearchimedeans7041  
Our website: https://archim.soc.srcf.net/
Our Facebook page:   / archimedeans  
Join our mailing list: https://lists.cam.ac.uk/mailman/listi...
Join our Microsoft Teams: http://forms.gle/tGHnN7XhEUVCeRgr6