Discover and read the best of Twitter Threads about #Graphs
Most recents (11)
The First-Time Machine Learning Playbook.
(Read this if you want to efficiently learn machine learning, avoid frustration from searching for resources, and build a career in ML.)
#Ship30For30 #ML #MachineLearning
(Read this if you want to efficiently learn machine learning, avoid frustration from searching for resources, and build a career in ML.)
#Ship30For30 #ML #MachineLearning
What people think you need:
Most people think you need to be a math and stats expert to learn ML. ML can be math intensive, but it’s not a barrier to entry.
What you need:
• The Ability to code
• An Open and Curious Mind
• A Good starting point
Which leads me to …
Most people think you need to be a math and stats expert to learn ML. ML can be math intensive, but it’s not a barrier to entry.
What you need:
• The Ability to code
• An Open and Curious Mind
• A Good starting point
Which leads me to …
Where to Start:
If you can code, then start here: course.fast.ai
The @fastdotai Course is a fantastic place to start. Instead of going from the basics towards an application, it starts from an application and breaks it down piece by piece.
Did I mention it’s free?
If you can code, then start here: course.fast.ai
The @fastdotai Course is a fantastic place to start. Instead of going from the basics towards an application, it starts from an application and breaks it down piece by piece.
Did I mention it’s free?
Want to create a Graph Neural Network based recommendation engine? You first have to get intimate with its performance metrics: recall@k and precision@k. Here are a few helpful illustrations I made that help visualize what you are measuring:
First, the definition of precision and recall.
- Precision: "What ratio of the top k elements that we recommended are relevant?"
- Recall: "What ratio of the existing pool of relevant items did we get right in the top k elements?"
In mathematical terms:
- Precision: "What ratio of the top k elements that we recommended are relevant?"
- Recall: "What ratio of the existing pool of relevant items did we get right in the top k elements?"
In mathematical terms:
Let's take an example to illustrate this, with simple tabular data first. We have a list of item ratings that we take as the baseline (targets) and we have the recommendation that our model spat out.
"From undirected to directed networks of dynamical agents"
Today's SFI Seminar from @robinus88 (@UCSantaBarbara), streaming now:
(Follow this 🧵 for highlights and select slides)
Today's SFI Seminar from @robinus88 (@UCSantaBarbara), streaming now:
(Follow this 🧵 for highlights and select slides)
"[This is] the main question when we talk about power grids...it could be water, it could be gas, it could be opinions transmitted over social media:"
- @robinus88 (@UCSantaBarbara), streaming now:
- @robinus88 (@UCSantaBarbara), streaming now:
"We want to keep the right-hand side of this equation as close to zero as possible. What happens if you produce too much, the frequency increases, which we don't want for a variety of reasons."
- @robinus88 (@UCSantaBarbara), streaming now:
#electricity
- @robinus88 (@UCSantaBarbara), streaming now:
#electricity
Daily Bookmarks to GAVNet 06/04/2021 greeneracresvaluenetwork.wordpress.com/2021/06/04/dai…
Uganda records highest single-day Covid cases
monitor.co.ug/uganda/news/na…
#uganda #COVID19 #CaseSpike #PandemicResponse
monitor.co.ug/uganda/news/na…
#uganda #COVID19 #CaseSpike #PandemicResponse
#MonkeySeeMonkeyDo
Extraction of Local World Dynamics (Newtonian Scale) from E2E Image Based Learning using Graphs
Q: Can #SkyNet Learn to Hit a Curve-Ball using Video from a Batter's View?
Extraction of Local World Dynamics (Newtonian Scale) from E2E Image Based Learning using Graphs
Q: Can #SkyNet Learn to Hit a Curve-Ball using Video from a Batter's View?
1/28: The complex web of relationships of the envisioned #Stakeholder #Capitalism is complex. It can only be holistically grasped by network-graphs. This is a breakdown-compilation of the 4th Industrial Revolution, the global Medical Industrial Complex & more.
EyeCandy-Thread⬇️
EyeCandy-Thread⬇️
2/28: This is a remake of a previous thread. The graphs were created via @twittlesis, but have been deleted now either due to censorship or other reasons.
Littlesis.org user 'WrenchInTheGears' created most of them, that's Alison McDowell (@Philly852).
Littlesis.org user 'WrenchInTheGears' created most of them, that's Alison McDowell (@Philly852).
3/28: Luckily I was able to backtrack most the relevant and lost graphs via WebArchive (I guess @Philly852 archived them). The archives are nevertheless interactive, you can click on the nodes, retrieve archived database entries.
Proud to announce our newest graph #research #paper, we introduce directional aggregations, generalize convolutional #neuralnetworks in #graphs and solve bottlenecks in GNNs 1/5
arxiv.org/abs/2010.02863
Authors:@Saro2000 @vincentmillions @pl219_Cambridge @williamleif @GabriCorso
arxiv.org/abs/2010.02863
Authors:@Saro2000 @vincentmillions @pl219_Cambridge @williamleif @GabriCorso
By using an underlying vector field F, we can define forward/backward directions and extend differential geometry to include directional smoothing and derivatives. By using different directional fields, the GNN aggregators become powerful enough to generalize CNNs. 2/5
We propose to use the gradient of the low-frequency eigenvector as directional vector field to guide the aggregation. We theoretically prove that it reduce both over-smoothing and over-squashing. 3/5
New paper: Tooling-up for infectious disease transmission modelling. The Intro to the #Epidemics special issue on Computational Methods for Modelling an Infectious Disease (COMMAND) doi-org.ez.lshtm.ac.uk/10.1016/j.epid… @GrahamMedley @EleanorMRees @N_R_Waterlow @cmmid_lshtm In the series:
1. Choices and trade-offs in inference with infectious disease models by @sbfnk and A King doi.org/10.1016/j.epid… #IDmodelling #modelchoice
2. Desirable BUGS in models of infectious diseases by @kathmoreilly @drrachellowe and others doi.org/10.1016/j.epid… #JAGS #graphs
“Why do companies ask data structure&algorithm questions? It’s not like you’d use this day to day...”
At my past 3 companies - Skype/Microsoft, Skyscanner, Uber - I needed to use them to write some code, and *especially* to understand things. Here’s some examples. A thread.
At my past 3 companies - Skype/Microsoft, Skyscanner, Uber - I needed to use them to write some code, and *especially* to understand things. Here’s some examples. A thread.
Real-time #economic and #market indicators will translate into a stunning series of (lagged) economic data points in the weeks to come, with Friday’s #JobsReport likely being a key first glimpse of what we’re in store for: eg. jobless claims 5X more than the next largest decline.
When all is said and done, the #unemployment rate might exceed the 2008 high, and the second quarter #GDP decline could be -10% (or worse), before we begin to recover.
Hi, @FrankPallone
I was one of the alleged bots #NotABot defamed for that @JoeSmyser @DrArturoNJ @NicholsonForNJ @PublicGoodProj @WSJ report.
Shall I expect an invite to @EnergyCommerce offices to sign in, in person, to my account?
We can also discuss manipulation of #data.
I was one of the alleged bots #NotABot defamed for that @JoeSmyser @DrArturoNJ @NicholsonForNJ @PublicGoodProj @WSJ report.
Shall I expect an invite to @EnergyCommerce offices to sign in, in person, to my account?
We can also discuss manipulation of #data.
@FrankPallone @JoeSmyser @DrArturoNJ @NicholsonForNJ @PublicGoodProj @WSJ @EnergyCommerce 👈#NotABot we can discuss allegations as to why I and other's data has been used to accuse innocent people of committing what I consider crimes @JoeSmyser @DrArturoNJ @NicholsonForNJ @PublicGoodProj @WSJ @EnergyCommerce like marketing to children.
@FrankPallone @JoeSmyser @DrArturoNJ @NicholsonForNJ @PublicGoodProj @WSJ @EnergyCommerce 👈#StillNotABot we can discuss how research fraud @JoeSmyser @DrArturoNJ @NicholsonForNJ @PublicGoodProj @WSJ @EnergyCommerce *may* have played a role in these hearings...
