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
If you take any complicated deep learning experimental setup, chances are you can remove a few modules (or replace some trained features with random ones) with no loss of performance. Get rid of the noise in the research process: do ablation studies.
Can't fully understand your system? Many moving parts? Want to make sure the reason it's working is really related to your hypothesis? Try removing stuff. Spend at least ~10% of your experimentation time on an honest effort to disprove your thesis.
Those two fun papers come too mind, when talking about "If I can't break it, I can't understand it" http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005268 … https://www.cell.com/cancer-cell/fulltext/S1535-6108(02)00133-2 …
classic randomized "Dropout" ? or more emphatic surgical removal of nodes?
Ablation studies are also how you can surprise yourself with a far simpler or more efficient implementation too :) Best of all, the faster your model runs, the quicker you can find such models as running many ablations is trivial ^_^
Don't wait until your model is already complex for ablations either. The best accident I ever made was trying to improve @PyTorch's tutorial language model as I wanted it to remain relatively simple and fast for beginners. I kept looking for low hanging fruit until it was ~SotA.
"old" but gold comment.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
