anilananth.bsky.social
Journalist with bylines in Nature, Quanta, Scientific American, New Scientist, and many more; former deputy news editor at New Scientist Author of 4 popular science books, including WHY MACHINES LEARN: The Elegant Math Behind Modern AI; TED speaker
36 posts
1,665 followers
1,012 following
Prolific Poster
Conversation Starter
comment in response to
post
Thank you Rohit
comment in response to
post
Happy birthday, Michael!
comment in response to
post
In good company with @anilananth.bsky.social's book!
comment in response to
post
Have added you to my new list Anil, thanks for flagging go.bsky.app/AFSGdUQ
comment in response to
post
Why Machines Learn: The Elegant Math Behind Modern AI, by Anil Ananthaswamy @anilananth.bsky.social . Writing a book about math that is compelling and includes enough math to actually give a feel for the topic is hard. This book mixes history and math in a way that I found quite compelling.
comment in response to
post
5/5 All this and more in WHY MACHINES LEARN, in the chapter The Algorithm That Put Paid To A Persistent Myth (the myth being the one propagated by Minsky and Papert that even MLPs could not solve the simple XOR problem).
US: penguinrandomhouse.com/books/677608...
UK: penguin.co.uk/books/446849...
comment in response to
post
4/5 There were many people between Rosenblatt on the one hand and Rumelhart, Hinton and Williams on the other, whose contributions were important to the development of the backpropagation algorithm and deep learning.
comment in response to
post
3/5 It took a series of small but significant tweaks to Rosenblatt's artificial neuron--for example tweaking the step thresholding function to something more continuous, such as a Sigmoid function-- and the use of the chain rule that finally led to the seminal Rumelhart, Hinton and Williams paper.
comment in response to
post
2/5 In his 1961 tome, The Principles of Neurodynamics, Rosenblatt clearly elucidates the idea that errors from the final layer will have to be "back propagated" -- he used that phrase-- to earlier layers, if such networks could be trained. He just didn't know how!
comment in response to
post
Why Machines Learn - The Elegant Math Behind Modern AI
By Anil Ananthaswamy @anilananth.bsky.social - a fantastic, accessible introduction to the math behind AI www.penguinrandomhouse.com/books/677608...