@Meaningness https://blog.keras.io/the-limitations-of-deep-learning.html … Reminds me of ur retweet of @The_Lagrangian, Scooby Doo unmasking deep learning as convex optimization
-
-
Currently reading the follow up post, also good: https://blog.keras.io/the-future-of-deep-learning.html …
-
Follow The Morning Paper? Write-ups of ML & cloud architecture papers. https://blog.acolyer.org
- 1 more reply
New conversation -
-
-
But isn't DL a very nonconvex problem?
-
DL is a solution, not a problem. Learns via stochastic gradient descent, and yes technically cost function is non-convex, however.. (1/n)
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.