Tweet by François Chollet
- Tweet
- Mar 20, 2021
- #Deeplearning #MachineLearning
Tweet
Deep learning excels at unlocking the creation of impressive early demos of new applications using very little development resources.
The part where it struggles is reaching the level of consistent usefulness and reliability required by production usage.
The part where it struggles is reaching the level of consistent usefulness and reliability required by production usage.
Mentions
See All
Balaji Srinivasan @balajis
·
Jul 3, 2021
Good thread on the last mile problem for deep learning. The main thing that’s worked in the successful production ML apps I’ve supervised (from variant calling to anti-fraud) is a cheap, fast feedback loop to turn new in-field training examples into online model updates.