Timothy Mann

I realized after all these years that the reason I haven't been eating pizza for breakfast was because it wasn't on a bagel.

3) Using model weight averaging (Izmailov et al., 2018) consistently provides a boost in robustness. In fact, weight averaging provides improvements as large as those provided by TRADES. 4) Tuning the mixing of additional unlabeled data (from Carmon et al., 2019) also helps.(4/4)

1) Combining TRADES (Zhang et al., 2019) with early stopping is superior to classical adversarial training (Madry et al., 2017; Rice et al., 2020) 2) Larger models and using SiLU/Swish provide improvements in clean and robust accuracy (both phenomena were already observed). (3/4)

Our findings generalize to other datasets (e.g., CIFAR-100) and other norms (e.g., L2). With additional data, we surpass the previous state-of-the-art by 6.34%. Without additional data, we improve on the results of 3 state-of-the-art methods that leveraged additional data. (2/4)

Have you ever wondered how to reach 65% robust accuracy on CIFAR-10 (against l-infinity norm-bounded adversaries of size 8/255)? In collaboration with Chongli Qin, @JonathanUesato, @realKingTim and @pushmeet, we are releasing our recipe at arxiv.org/abs/2010.03593. (1/4)

Exciting results from our partnership with @googlemaps! Using advanced Graph Neural Networks we were able to improve the accuracy of ETAs in major cities across the world by up to 50% bit.ly/2Z0IMYC

What's up with people saying they hate cliches. They are popular for a reason. I love cliches.

Zetus Lapetus! Watching Zenon girl of the 21st century with my children and cringing.

"Merriam-Webster even reported that lookups for the word 'pettifogger' spiked 30,800% ." -BBC

If you haven't listened to the soundtrack from Quake 2 prepare to have your 1997 mind blown: music.youtube.com/watch?v=jY6yBT…

Training data is often collected through a biased process. Models trained on such data are inherently biased. We demonstrate how adversarial training through disentangled representations can reduce the effect of spurious correlations present in datasets: arxiv.org/abs/1912.03192

Can we just stop with the unnecessary chocolate?

Still seeing unauthorized soliciting on my morning train. When are you going to step up security? @TLRailUK

Another day; another unauthorized passenger selling merchandise on the train. #thameslink @TLRailUK Please step up security. Thank you.

#thameslink if you can't handle minor security infractions, how can we trust you with more serious security issues?

#BoycottNRA EOM

@startrekcbs Discovery isn't Star Trek! I stopped watching at episode 5. It's just bad.

Sweaters By Sam Barsky - theinspiration.com/2017/11/sweate…

Banana milk! It sounds delicious, but it's oh so disappointing.

Merry Christmas everybody!