Justin Domke (@JustinDomke) header
Justin Domke (@JustinDomke) profile picture

Justin Domke

@JustinDomke

Fan of high nat/flop ratios.

justindomke.wordpress.com
Joined May 2020
  • Tweets42
  • Following62
  • Followers347
10Media
Photo by Justin Domke (@JustinDomke)Photo by Justin Domke (@JustinDomke)Photo by Justin Domke (@JustinDomke)Photo by Justin Domke (@JustinDomke)Photo by Justin Domke (@JustinDomke)Photo by Justin Domke (@JustinDomke)Photo by Justin Domke (@JustinDomke)Photo by Justin Domke (@JustinDomke)Photo by Justin Domke (@JustinDomke)Photo by Justin Domke (@JustinDomke)
So I wrote a library to do Bayesian inference using a notation as close to mathematical notation as possible: github.com/justindomke/pango…
Image from Justin Domke (@JustinDomke)
Image from Justin Domke (@JustinDomke)
1
5
59
8,603
Excited to share this retrospective of papers shown at ML conferences in 2021 by folks from @umasscs including several #NeurIPS2021 papers: all.cs.umass.edu/ML2021.html #UMassML21
13
1
26
Hark, Bayesians! Everyone's favorite approximate inference symposium is back! The submission deadline is November 19th, so send us all your recent work on approximate Bayesian inference. Please share, retweet, and tell all your friends: approximateinference.org/cal…
1
21
5
72
justindomke.wordpress.com/20…
Is anchoring a reliable cognitive bias?

Is anchoring a reliable cognitive bias?

Say you get a large group of people and split them into three groups. The first group gets this question: What is the length of the Mississippi river? The second group gets these questions: Is the …

justindomke.wordpress.com
4
Turning confidence sets into confidence intervals. (Or how you could accidentally invent the clopper-pearson interval without math using visualization and greedy choices.) justindomke.wordpress.com/20…
Image from Justin Domke (@JustinDomke)
Image from Justin Domke (@JustinDomke)
Image from Justin Domke (@JustinDomke)
1
4
my attempt to *really* explain confidence intervals without math by discretizing everything justindomke.wordpress.com/20…
Image from Justin Domke (@JustinDomke)
Image from Justin Domke (@JustinDomke)
Image from Justin Domke (@JustinDomke)
1
1
10
The replication game: try to predict which psychological effects replicate justindomke.wordpress.com/20…
Image from Justin Domke (@JustinDomke)
Image from Justin Domke (@JustinDomke)
Image from Justin Domke (@JustinDomke)
2
gaussian process regression vs. printing out your data and drawing a bunch of curves on top of it by hand justindomke.wordpress.com/20…
Image from Justin Domke (@JustinDomke)
Image from Justin Domke (@JustinDomke)
Image from Justin Domke (@JustinDomke)
9
55
18
225
Just finished teaching a graphical models course. This was definitely my favorite result. ("Look at probabilities and utilities working together in harmony!") But somehow, the proof is too easy to be satisfying...
Image from Justin Domke (@JustinDomke)
1
10
I think this paper is basically doing distillation of a model learned by a dog's brain into a machine learning model. (What's the inductive bias that dogs use to generalize so well?) journals.plos.org/plosone/ar…
Image from Justin Domke (@JustinDomke)
1
Justin Domke retweeted
AABI (@aabi_org) profile picture
AABI@aabi_org20 Jan 2021
In the third #AABI2020 session on Wednesday, Jan 27, Justin Domke will present "Some Embarrassing Questions about Variational Inference". Register at us02web.zoom.us/webinar/regi… or watch our Youtube Livestream at youtube.com/channel/UCZLWLfZ…
2
11
1
38
The Flat Earth Society has disconcertingly good information on Symplectic Integrators. wiki.tfes.org/Symplectic_Int…
3
For general Bayesian inference, how good is: Laplace init? Meh IW training? Meh IW sampling? Good STL gradient estimator? Good Normalizing flows: Really good Careful step size? Really, really, really good proceedings.neurips.cc/paper… #NeurIPS2020
Image from Justin Domke (@JustinDomke)
1
5
Want faster variational inference? You can use a Taylor expansion, but this only works for mean-field. We tried something "dumb" for other dists: Just optimize a quadratic online. Surprisingly, this might be better anyway. #NeurIPS2020 papers.nips.cc/paper/2020/fi…
Image from Justin Domke (@JustinDomke)
1
4
Calculating variance in the log domain justindomke.wordpress.com/20… (I'm sure a better solution than this was known decades ago... let me know if you know what it is!)
Calculating variance in the log domain

Calculating variance in the log domain

Say you’ve got a positive dataset and you want to calculate the variance. However, the numbers in your dataset are huge, so huge you need to represent them in the log-domain. How do you compu…

justindomke.wordpress.com
Back to Basics: Approximate Bayesian Computation -- justindomke.wordpress.com/20…
Back to Basics: Approximate Bayesian Computation

Back to Basics: Approximate Bayesian Computation

(All based on these excellent slides from Umberto Picchini) “Approximate Bayesian Computation” sounds like a broad class of methods that would potentially include things like message pa…

justindomke.wordpress.com
3
Load more