It's a rather bizarre perspective, like a bicycle rental shop saying "it's feasible to go from Paris to Beijing on our bicycles, but it would take very long". You just need a faster vehicle. If you can't do something with a small dataset, that's a technology problem.
-
-
Show this thread
-
Every problem is solvable with k-NN. You "just" need enough data.
Show this thread
End of conversation
New conversation -
-
-
Yes and no. For vision, we can use gan for image data augmentation and get bigger dataset because there are patterns and vision is predictable. How constantly evolving persons think in uncertain situations, not so much. The issue is lack of understanding human decision making
-
The more we understand the more ways we find to do same thing with more accuracy and efficiency and this is how tech develops ... I guess that's the very point of post as well Developing better tech to solve a problem with smaller datasets !
- Show replies
New conversation -
-
-
Sometime you need a paradigm shift within the same discipline. For instance it's sometime impossible to collect enough perfect pairs of data to train a model. But approach this with GAN or VAE, and a large unpaired dataset will solve your problem.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Are generative datasets infinite??
-
Can we dream to the stars?
End of conversation
New conversation -
-
-
We could teach a child to read, but they will need to live at least 4-6 years. Seems like a sensible thing to say to me!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
They won't abandon the golden hammer. I like thathttps://twitter.com/trylks/status/1084165440656551936 …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
If the value created by X is (maybe) superior to the cost of the dataset, then it is a good slide for a startup :)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.