@_amiller_

Research Scientist @apple. ML/Stats & Health.

andymiller.github.io
Joined September 2010
  • Tweets358
  • Following436
  • Followers779
19Media
Andy Miller retweeted
Diana Cai
@dianarycai6 Dec 2018
1/ Come to the Bayesian nonparametrics workshop at #NeurIPS2018 on *Friday* in Room 517D! We have a great set of speakers, panelists, poster sessions, lunch+TensorFlow Probability tutorial! Schedule: sites.google.com/view/nipsbn… Accepted papers: sites.google.com/view/nipsbn… Highlights:

BNP@NeurIPS 2018 - Schedule

Workshop schedule Invited speaker abstracts and bios available at the end. Location: Palais des Congrès de Montréal, Montréal CANADA - Room 517 D Session 1: 8:25-8:30 Introductions 8:30-9:00 Invited...

sites.google.com
1
3
1
16
Andy Miller retweeted
Ethan Plaut@ethanplaut1 Jul 2018
This is not how AI works, or how writing works, or how education works. xitter.cf/NPR/status/10132266816…
NPR
@NPR1 Jul 2018
Computers already drive cars and detect cancer, so they can certainly handle grading students' essays, developers say. n.pr/2tGqBYv
19
136
10
638
Bias laundering: when our biases no longer just live in our minds, but get baked into algorithms (e.g. via historically biased data), they become harder to acknowledge, and harder still to identify and fix.
23
219
23
529
pretty cool data viz - detecting fake followers on twitter (unrelated, please don't look at my followers) nytimes.com/interactive/2018…
7
Replying to @StatMLPapers
TL;DR: There are two problems: The best Gaussian might still be a bad approximate posterior, and the recognition net might put the Gaussian in a bad place. It turns out the first problem isn't so bad, because the generator can adapt to it. Kudos to @_chriscremer and @lxuechen!
1
11
Estimate+predict Bayesianly. Use data efficiently; propagate error coherently; generate precise hypotheses. Validate+falsify frequentist-ly (and frequently!). Run experiments that reject those hypotheses with high probability if they are wrong. Learn both paradigms. Alternate.
1
20
2
58
I wrote a gentle INLA tutorial, if you've been wanting to learn the gist / get started with INLA, it might be a good place to begin. Happy holidays! #rstats #inla #bayes precision-analytics.ca/blog-…
8
74
6
207
Three challenges in our field Hype -- Don't oversell. Talk about what you know. Ethics -- Be aware of the potential for evil in applications of new tech. Do the right thing. Science -- Empiricism is a valid way to produce new knowledge. So is theory. Hacks & obfuscation aren't.
5
274
16
691
Andy Miller retweeted
Rembert Browne@rembert5 Dec 2017
Here’s the thing, Sane person from Atlanta. I know you don’t like Kasim. It makes sense. Duh. But don’t let that make you so basic. That you vote for @marynorwood. Tomorrow: Vote for @KeishaBottoms.
2
45
4
94
It’s hard to miss #NIPS2017 just because I am an Iranian. I always thought politics should have a limit, but apparently not this time!
1
23
3
74
Best-ever algorithm? This must be what everyone is talking about at #NIPS2017 #FreeLunch 💸🌳📈 xitter.cf/quantamagazine/status/…
Quanta Magazine
@QuantaMagazine4 Dec 2017
A more effective way of tracking trending topics: buff.ly/2BtUwWb
5
Time for Mass. to adopt automatic voter registration, @PatrickRoath patriotledger.com/opinion/20…
7
For folks who don't like clicking thru, here's the abstract. Overlap in Observational Studies with High-Dimensional Covariates. arxiv.org/abs/1711.02582.
3
5
2
16
Andy Miller retweeted
Vox
@voxdotcom7 Nov 2017
The GOP tax bill could be a disaster for PhD students vox.com/policy-and-politics/…

The GOP tax bill could be a disaster for PhD students

“It would really upend the American PhD system,” a professor says.

vox.com
4
95
45
82
When I tell my computer science colleagues that there are so many fairness definitions, they are often surprised and/or confused. [Thread] xitter.cf/random_walker/status/9…
Arvind Narayanan
@random_walker24 Oct 2017
Today in Fairness in Machine Learning class: a comparison of 21(!) definitions of bias and fairness. Notes coming soon.
6
112
23
258
Midnight tomorrow at the latest time zone. :-) (retweeting as a PSA!) xitter.cf/yeewhye/status/9254264…
Yee Whye Teh@yeewhye31 Oct 2017
Replying to @dustinvtran
@dustinvtran What's the AABI2017 deadline time? We can't find it on either easychair, or the call for papers. Thanks.
1
6
Analysts like the Dodgers' @dcervone0, PhD '15 in Statistics, have ushered in a new era of baseball buff.ly/2xGFBWm #worldseries
11
1
21
Figuring out *how* to optimize something is mostly a math problem. But figuring out *what* to optimize is mostly not.
47
619
85
1,781
👇 xitter.cf/alexdamour/status/9203…
Alexander D'Amour (@alexdamour.bsky.social)@alexdamour17 Oct 2017
Great read. Justices need more stat education, but presentation of stat args can also be improved. 1/n xitter.cf/FiveThirtyEight/status…
4
Andy Miller retweeted
Luke Bornn@LukeBornn10 Oct 2017
Simpson's Paradox in Basketball: Shooting improves as defenders guard closer (!!). Must condition on shot location. h/t Nate + @afranks53
7
68
12
202
Load more