Great papers on ML design?
Hey folks, I have been preparing for system designs and read a couple of papers I think are great and helped my preparation a lot. For instance GFS paper Dynamo and Cassandra paper Facebook TAO paper I’m starting to prepare ML designs now and would love to do the same. Is there any industry papers that you’d recommend for practical model arch design, ML system design and feature engineering? One example I found was YouTube’s 2016 paper abt their recommender system. But I’m sure there are many out there like Uber’s time estimation etc Thanks! TC: 0 since I left Elon #layoff #software #google #meta #microsoft #twitter #machinelearning #systemdesign #interview
Following, I would like to know some too.
OP, interested in a job ? Are you a generalist or specialist ?
Did a lot of work on ML infra side. Yeah why not :)
Latent Dirichlet Allocation https://www.jmlr.org/papers/volume3/blei03a/blei03a.pdf Written by Michael Jordan himself.
Thanks Google! LDA is cool I’m not sure who’s gonna ask this in interview though. Any other industry paper?
“Rome was built in a day” or something like that. Forgot the actual name
Get Chip Huyens book and read her blog.
You might be interested in: machinelearningatscale.com :)
There are many cool blogs and GitHub repos out there, try gsearching them! NB nothing beats true experience
Check the full stack deep learning course from UCB
Uber and Spotify have great blogs describing their ML system as well.
Hey preparing for ML designs too want to hit me up on dm?
Sure lol