Most people I meet overestimate what deep learning can do (it's curve-fitting, don't expect it to do discrete symbol manipulation, it will solve symbolic tasks via embedding + interpolation) and simultaneously underestimate what you can do with curve-fitting given enough data
-
Show this thread
-
Replying to @fchollet
Extracting information that isn’t there is a current deep learning mania
2 replies 2 retweets 11 likes -
Replying to @KarlLandheer
Or embedding one-off discrete objects on a curve with no chance of generalization (i.e. most Deep RL). Lots of people are playing with the tech with no idea about the first principles that make it work
2 replies 4 retweets 40 likes -
Replying to @fchollet @KarlLandheer
Are this principles present in deep learning with python? I got the MEAP and I'm going through it
1 reply 0 retweets 5 likes
Replying to @__mvalente @KarlLandheer
Yes, chapters 5 and 14 for the most part, but generally the entire book
5:20 PM - 20 Jul 2021
0 replies
0 retweets
8 likes
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.