DeepLearningAI
Take the Deep Learning Specialization: http://bit.ly/2vBG4xl
Check out all our courses: https://www.deeplearning.ai
Subscribe to The Batch, our weekly newsletter: https://www.deeplearning.ai/thebatch
Follow us:
Twitter: https://twitter.com/deeplearningai_
Facebook: https://www.facebook.com/deeplearningHQ/
Linkedin: https://www.linkedin.com/company/deeplearningai
Source
Why did you erase the squared at 2:46? Shouldn't RMSprop have a squared term for the bias as well?
This nailed down the Adam paper. Thanks alot
You are so sweet. Thank you Sir, for these awesome videos!
you really dont think that statement of the problem that ADAM solves is of relevance, when you are introducing ADAM?
any time I want to implement ML from scratch, I watch all Andrew's videos from beginning to end! I don't know how to express my appreciation to this great man.
You are my god.
Roasting at the end ! Hahaha
Eve Optimization Algorithm will come soon!
This video is closely related to the video "Bias Correction of Exponentially Weighted Averages". Please revisit that video if you feel this is too confusing.
比助教講得好太多了
great explntion.Meed to watch again
Clarification about Adam Optimization
Please note that at 2:44, the Sdb equation is correct. However, from 2:48 , the db² lost the ².
The bottom right equation should still be:
Sdb = β₂Sdb + (1 – β₂)db²
what is s and v
Only understood his friend has nothing to do with Adam optimization!
Ow my ears
Could anyone give me a list of the notations he mentions in the video or direct me towards a video that has those explained? Main issue with understanding the concept in the video is the lack of explanation of the notations used.
First task of significance is for me to figure out how to spell Andrews last name then I move on to the algorithm 🤓
😂 6:34
what is t I do not completely understand
The very best and most succinct explanation of ADAM I've ever seen. Things become crystal clear if one watches L06 to L08 in a row.
this man is a Legend!!
Hey there I know I am late to the party but I have a pressing question the rest of the internet has failed to answer so far.
I currently have to work with a model and network I didn't design and my job is to basically find out whats wrong so naturally I need to understand the LOC used.
There was a line I havent found any example for: optimizer = keras.optimizers.Adam(0.002, 0.5)
I am still studying so I am not that well versed in Keras or anything AI so far really but I wanna know if this second value refers to the beta_1 or any other value I am not noticing.
The documentation has me puzzled so far so I hope theres someone here who can answer this.
-1 no knowledge about why Adam works better then previous algorithms is provided
It would be easier if you just typed instead of handwrite I can’t read it