on compressing it down to one variable (slider). human in the loop machine learning has lot to offer in this space.pic.twitter.com/qu9YG7WKlH
You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
on compressing it down to one variable (slider). human in the loop machine learning has lot to offer in this space.pic.twitter.com/qu9YG7WKlH
Why do gradient descent when you can ask a user to do it for you :)pic.twitter.com/uhUG7YajjG
automating things in general is hard but most often what is desirable to have is a way to fix the errors or unwanted outcomes based on user's feedback so there's an active loop between the tool and the user.
This is creative! On average, how many times user need to manually search for model convergence?
And is this framework generally applicable to most GAN learning? (e.g. colorization, image translation ... etc)
This is interesting. Decomposing J is something we were thinking about for uncertainty estimation.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.