Montgomery Flora, Ph.D. (@MontePhD) header
Montgomery Flora, Ph.D. (@MontePhD) profile picture

Montgomery Flora, Ph.D.

@MontePhD

Self-motivated individual actively pursuing knowledge. Passions include being a loving husband/father, studying severe weather, and playing music.

 Cumming, GA
monteflora.wixsite.com/mflora
Joined January 2019
  • Tweets278
  • Following105
  • Followers159
68Media
Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)Photo by Montgomery Flora, Ph.D. (@MontePhD)
Montgomery Flora, Ph.D. retweeted
Corey K Potvin (@corey_potvin) profile picture
Corey K Potvin@corey_potvin25 Jul 2024
Our paper combining AI/ML, storm-scale ensemble prediction, and "forecasting forecast skill" is now in final form in @AMS_AIES: doi.org/10.1175/AIES-D-23-01… Co-authors: @MontePhD, Patrick Skinner, @tonyreinhart, @BrianWxFLA
1
4
371
Montgomery Flora, Ph.D. retweeted
Harold Brooks (@hebrooks87) profile picture
Harold Brooks@hebrooks8726 Mar 2024
Now online at @AMS_AIES, a paper on the variety of names given to the same quantities from the 2x2 table. "A rose by any other name...." doi.org/10.1175/AIES-D-23-01…
1
5
664
Montgomery Flora, Ph.D. retweeted
CIWRO (@CIWRO_) profile picture
CIWRO@CIWRO_1 Feb 2024
Severe weather forecasts are impeded by the limited NWP guidance at short lead times. Machine learning, applied to @NOAANSSL's Warn-on-Forecast System, can inform forecasts of wind, hail and tornado potential at watch-to-warning lead times. #AMS2024
Image from CIWRO (@CIWRO_)
6
1
30
3,794
Montgomery Flora, Ph.D. retweeted
NOAA Hazardous Weather Testbed (@NOAA_HWT) profile picture
NOAA Hazardous Weather Testbed@NOAA_HWT25 Jan 2024
We are looking for broadcast meteorologists and emergency managers interested in talking about possible changes to severe weather outlooks! Click here to learn more and to apply: hwt.nssl.noaa.gov/docs/EM_br…
3
28
5
42
19,156
Montgomery Flora, Ph.D. retweeted
MrBeast (@MrBeast) profile picture
MrBeast
@MrBeast22 Jan 2024
I’m gonna give 10 random people that repost this and follow me $25,000 for fun (the $250,000 my X video made) I’ll pick the winners in 72 hours
368,305
2,481,284
219,990
1,828,380
286,867,104
How do researchers define interpretability and explainability? An overview: buff.ly/476R81x Summary: No consensus My opinion: It's already the Wild West, no one will stick to any definition. Most pragmatic is to treat them interchangeably.
Table 1 from the paper linked in the post

ALT Table 1 from the paper linked in the post

16
63
6
279
44,233
Montgomery Flora, Ph.D. retweeted
CIWRO (@CIWRO_) profile picture
CIWRO@CIWRO_10 Jan 2023
Chad Wiley: Combining deep learning (DL) with current radar in the Warn-on-Forecast System enables skillful forecasts of severe thunderstorm locations. Multi-Radar/Multi-Sensor System (MRMS) data will serve as ground truth for evaluation of the DL model. #AMS2023
Image from CIWRO (@CIWRO_)
1
3
483
The experimental @NOAA Warn-on-Forecast System aims to increase lead times for tornado, severe thunderstorm, & flash flood warnings via rapid-update, storm-scale ensemble modeling. WoFS is running today and forecasts will begin at ~12:00 PM CT. go.usa.gov/xu8NC
1
14
3
40
Montgomery Flora, Ph.D. retweeted
𝗝𝗼𝘀𝗵𝘂𝗮 (@Joshua_Bambrick) profile picture
𝗝𝗼𝘀𝗵𝘂𝗮@Joshua_Bambrick11 Mar 2022
Re-stumbled across this diagram from @rasbt covering the most popular techniques for improving the generalisability of deep neural nets. A really excellent visualisation for ML practitioners.
Sebastian Raschka (@rasbt) profile pictureSebastian Raschka
@rasbt10 Mar 2021
Replying to @rasbt
Thanks for the feedback, everyone! Below is a slightly extended version based on all the awesome responses I received. Will probably extend it some time with more concrete methods for some of the categories some time later this semester :)
Image from Sebastian Raschka (@rasbt)
1
8
29
Montgomery Flora, Ph.D. retweeted
Corey K Potvin (@corey_potvin) profile picture
Corey K Potvin@corey_potvin2 Feb 2022
Now in final form at doi.org/10.1175/WAF-D-21-007…
Corey K Potvin (@corey_potvin) profile pictureCorey K Potvin@corey_potvin4 Jan 2022
New WAF EOR by Will Miller: “Exploring the Usefulness of Downscaling Free Forecasts from the Warn-on-Forecast System”. doi.org/10.1175/WAF-D-21-007…
1
2
Montgomery Flora, Ph.D. retweeted
San Francisco 49ers (@49ers) profile picture
San Francisco 49ers
@49ers23 Jan 2022
WHAT A HEADS UP PLAY BY THE ROOK @TalanoaHufanga 🤙 #SFvsGB on FOX/49ers App
192
2,112
534
9,471
Montgomery Flora, Ph.D. retweeted
David Lombardi (@LombardiHimself) profile picture
David Lombardi
@LombardiHimself18 Jan 2022
Road teams went 1-5 in this weekend of the NFL playoffs. The one win belongs to the 49ers
31
331
38
4,321
Montgomery Flora, Ph.D. retweeted
San Francisco 49ers (@49ers) profile picture
San Francisco 49ers
@49ers17 Jan 2022
How Bout Them Niners!!!!
497
6,363
1,068
18,966
I have been thinking a lot about this paper about why explainable AI is not ready for patient-level explainability of AI in healthcare: thelancet.com/journals/landi… While I would not go as far as calling it a "false hope", there is food for thought in there. 1/n
The false hope of current approaches to explainable artificial intelligence in health care

The false hope of current approaches to explainable artificial intelligence in health care

The black-box nature of current artificial intelligence (AI) has caused some to question whether AI must be explainable to be used in high-stakes scenarios such as medicine. It has been argued that...

thelancet.com
2
16
4
41
Montgomery Flora, Ph.D. retweeted
Victo☈ Gensini, PhD, CCM (@gensiniwx) profile picture
Victo☈ Gensini, PhD, CCM@gensiniwx15 Oct 2021
I had a lot of fun working on this one. Now avail. via EOR via AMS WAF: "Machine learning classification of significant tornadoes and hail in the U.S. using ERA5 proximity soundings" doi.org/10.1175/WAF-D-21-005… We use game theory to help describe decision making by the ML model.
Image from Victo☈ Gensini, PhD, CCM (@gensiniwx)
Image from Victo☈ Gensini, PhD, CCM (@gensiniwx)
Image from Victo☈ Gensini, PhD, CCM (@gensiniwx)
1
14
4
89
I just published Functional Decomposition, a new chapter of the IML book 🥳🥳🥳 christophm.github.io/interpr… Functional decomposition is a fundamental but underappreciated concept of machine learning interpretability. It is one of the most central chapters in the book. A thread
4
11
45
Montgomery Flora, Ph.D. retweeted
Leonard Smith (@lynyrdsmyth) profile picture
Leonard Smith@lynyrdsmyth12 Sep 2021
One cannot evaluate a probability forecasting system given only a single forecast-outcome pair.
3
2
8
Our paper on General Pitfalls of Model-Agnostic Interpretation Methods for Machine Learning Models is on arxiv! 🎉🎉🎉 arxiv.org/abs/2007.04131 These pitfalls can really mess up your model interpretation. Our paper can help you avoid them. A thread. 1/n 🧵
General Pitfalls of Model-Agnostic Interpretation Methods for...

General Pitfalls of Model-Agnostic Interpretation Methods for...

An increasing number of model-agnostic interpretation techniques for machine learning (ML) models such as partial dependence plots (PDP), permutation feature importance (PFI) and Shapley values...

arxiv.org
7
83
11
319
Montgomery Flora, Ph.D. retweeted
Stephan Hoyer (@shoyer) profile picture
Stephan Hoyer
@shoyer20 Jul 2021
Can machine learning improve physics constrained optimization tasks, like those at the heart of numerical weather forecasting? Happy to share some our work on "Variational Data Assimilation with a Learned Inverse Observation Operator", tomorrow/today at ICML. 🧵
6
47
3
281
Montgomery Flora, Ph.D. retweeted
Roman Hornung (@HornungRoman) profile picture
Roman Hornung@HornungRoman16 Apr 2021
Thank you @BoulesteixLaure! Proud to present our preprint on the detection of interpretable interaction effects using our approach 'interaction forests' and its 'effect importance measure'. Eager to apply it? :) Check out our R package 'diversityForest' as well. #MachineLearning
Anne-Laure Boulesteix (@BoulesteixLaure) profile pictureAnne-Laure Boulesteix@BoulesteixLaure7 Apr 2021
New preprint with Roman Hornung on the detection of interaction effects using a new variant of the machine learning method „Random Forest“ (RF): epub.ub.uni-muenchen.de/7543… The last step of a long story. A thread (1/n)
3
6
Load more