Discover and read the best of Twitter Threads about #ML

Most recents (24)

Proto 6/10: Copy printed text to desktop with AR+ML

Code: github.com/cyrildiagne/ar…
Book: Neurones, les intelligences simulées, Frédéric Migayroux & al (Editions Hyx 2018 @centrepompidou)

#ML #AR #AI #AIUX #ARCore #ARKit #WebXR

Technical Insights: ↓
The magic here is to use ARCore + AugmentedImages rather than SIFT.
Phone gets a new desktop screenshot on touch and adds it to ARCore (< 100ms).
Tracking is crazy fast & precise.

Interesting alternative to touch screen for interactive installations!
The text detection is performed on device with @Firebase #MLKit. Super fast, good accuracy and cross platform.
Read 6 tweets
I take pride in being able to make complex things seem simple.

I've been learning machine learning actively, passively & actively again for the past 30 months.

And I'd like to share:

bit.ly/ml-smp-101

#machinelearning #learningmachinelearning #learningml #ml #ai

1/n
So I've decided to do something slightly uncomfortable. I'm going to put myself out there. A lot.

Blog posts, webinar series, talks (post-COVID-19) etc.

"Learning Machine Learning".

I love to debate (or argue)

#machinelearning #learningmachinelearning #learningml #ml #ai

2/n
In a recent very frustrating argument, I promised the audience that I would be back.

Yesterday, I posted the following on the WhatsApp group:
-
#machinelearning #learningmachinelearning #learningml #ml #ai

3/n
Read 22 tweets
What's the difference between a Data Engineer , a Data Analyst & a Data Scientist?

If you want to make sure you don’t lose your job in the next five years, you probably want to know something about Big Data, or even switch to a data-related career.
#Thread
#DataScience #Data
You don’t need to be an excellent statistician or a high-class mathematician to work in data science or analytics, nor do you need a lot of prior programming knowledge.
- Dr. Rebecca Pope (Head, Data Science d Engineering at KPMG)

#DataScience #Data #DataAnalyst #Statistics
However, you do need an interest in statistics, you do need to be willing to learn how to code, and you do need to know how to do some high level mathematical operations. Data scientists are not just statisticians.

#DataScience #Data #DataAnalyst #Statistics
Read 12 tweets
4/10 - Cut & paste your surroundings to Photoshop

Code: github.com/cyrildiagne/ar…

Book: @HOLOmagazine
Garment: SS17 by @thekarentopacio
Type: Sainte Colombe by @MinetYoann @ProductionType
Technical Insights: ↓

#ML #AR #AI #AIUX #Adobe #Photoshop
The secret sauce here is BASNet (Qin et al, CVPR 2019) for salient object detection and background removal.

The accuracy and range of this model are stunning and there are many nice use cases so I packaged it as a micro-service / docker image: github.com/cyrildiagne/ba…
And again, the OpenCV SIFT trick to find where the phone is pointing at the screen.

I also packaged it as a small python library: github.com/cyrildiagne/sc…

Send a camera image + a screenshot and you get accurate x, y screen coordinates!
Read 5 tweets
#ECUADORPOSTCOVID19
En un post anterior mencionaba que creo que Ecuador debe enfocarse hacia el desarrollo agrícola, y tener como objetivo que en 20 años seamos la más grande potencia agrícola del s. XXI, en función de los ratios de producción y márgenes o ganancia.
@CITECec
Para conseguir un #ECUADORPOTENCIAAGRICOLA debemos hacer cambios sociales, políticos, culturales, empresariales y tecnológicos.
De esto último tratarán éste y los hilos futuros, de cómo usar las nuevas tecnologías para desarrollar el sector agrícola del país.
@Construpositivo
El objetivo es incentivar a los empresarios, gobierno, inversionistas, hacia la explotación de estos nuevos recursos.
Escenario:
En 2050, se necesitará un 50% más de alimentos en el mundo.
Ver: bit.ly/3euCxSX
(FAO - La agricultura mundial en la perspectiva del año 2050)
Read 10 tweets
DAY 1: Python Installation on Windows.

Let start our 30 days journey of Twitter Python Training with:
* How to Install Python.
*Likely issues to faced after the installation of Python.

#Thread
👇👇👇
#TwitterPython #Python #ML #DL #AI #DataScience
I will like to talk about two ways of Installing and running Python on your personal computer and the method I 'prefer' out of the two.

1. Direct Installation from Python Website
2. Installation from Anaconda Website.

#TwitterPython #Python #ML #DL #AI #DataScience
1st Method:
Direct Installation from Python Website

Installing Python on your PC is not a difficult task.
It involves just a few simple steps:
a. Download binaries from python.org
b. Install the binaries
c Add Python to system environment variables
d. Install pip
Read 9 tweets
From a recent @martin_casado article posted on @a16z:

"Cloud infrastructure is a substantial – and sometimes hidden – cost for AI companies".

I'm sharing the techniques we use at @FloydHub_ to reduce this cost on #AWS and improve our gross margins [Thread] #ML #AI
0/ Not all AWS regions are priced the same. GPUs can be up to 90% more expensive across regions. Other than cost consider these when picking your AWS region: proximity to your geographical location, compliance requirements, and integration with any existing AWS infrastructure.
1/ Reserve your GPU instances and/or purchase Savings Plans. Review your GPU usage for the last 3-6 months and purchase 1-year plans based on that. Gives you 25-30% savings on your GPU bill.
Read 12 tweets
And the #dadlihack pitches have begun! First up, Team #EEQUALSMC2 and their solution to sweep the sea floor clean. Image
Delighted with the turnout at #Absip on beautiful #Antigua Image
Now pitching at #dadlihack it's team The Waves and their Corral2020 project looking at applying #AI to map building applications to existing standards, in order to increase overall resilience. Image
Read 18 tweets
Having reviewed a bunch of radiology papers lately with the same flaws - some #ML but most not - here are my top 10 tips! #AcademicChatter #radiology #AI
1/ Start with what is already known, where is the gap in knowledge and why it's important. Some papers are doomed from the start because there is no clear impact of the work. Remember the reviewer may not be super-specialised in your research niche.
2/ I need to understand the experimental design quickly, so a flowchart is really helpful. Put yourself in the shoes of a first-time reader. If the design is easy to follow then everything else fits into place.
Read 11 tweets
1/13 Hoje irei falar para vocês sobre dois conceitos que são menos independentes do que parecem para alguns: Causalidade e predição. Confesso que é até estranho tratá-las como duas coisas separadas, ou diferentes, e espero convencê-los ao... #IA #AI #ML #causality #bookofwhy
2/13 final dessa thread de que essa visão é fundamentada. Na década de 50, Jacob Yerushalmy realizou um estudo onde acompanhou 15 mil crianças da região da baía de São Francisco. Para surpresa de Yerushalmy, e contrariando o que já se mostrava forte na época (que fumar
3/13 fazia mal a saúde), seus resultados indicavam que bebês de mães fumantes nascidos com baixo peso tinham mais chances de sobreviver do que bebês de mães não fumantes nascidos com baixo peso. Não era um estudo de inferência causal, era apenas predição, alguns podem dizer.
Read 14 tweets
How To Train Interpretable Neural Networks That Accurately Extrapolate From Small Data. Today we released a new paper that showcases how to do just that using Scientific Machine Learning (#sciml) techniques to encode non-data scientific information.

stochasticlifestyle.com/how-to-train-i…
Our approach builds upon the work of @DavidDuvenaud but identifies that the differential equations one works with does not have to be a blackbox, but instead can utilize all of the available scientific models to encode as much prior information as possible.
@DavidDuvenaud This is a structure that we call the Universal Differential Equation: a differential equation with embedded universal approximators. Sometimes neural networks, sometimes Chebyshev polynomials, for us it really doesn't matter because they are small approximators.
Read 12 tweets
Starting 2020 off right by getting the benefit of hindsight & reading some classic papers.

#epielliereads
1st up, Take the Con out of Econometrics by Edward Leamer, 1983: all about pros & cons of randomized & observational studies plus answers to many of recurrent #econtwitter, #statstwitter, & #epitwitter arguments & even why #ML can’t do causal inference!
jstor.org/stable/1803924
In Part I randomized experiments vs natural experiments, aka observational studies, Leamer provides a nice summary of some of the problems that can arise in RCTs including that randomization guarantees validity *on average* but that chance imbalance can make any given RCT biased
Read 13 tweets
#AI #DevOps #Hackathon reminder. The final mile in any applied #ML project is deployment into the hands of the people that need it. This is an engineering problem requiring a specific set of skills which this hackathon will test. 1-4 people/team. Is your team ready for this? 1/n
Interested? Send a structured email to DevOpsAiChallenge@retina-ai.com with header "AI DevOps." Email should contain:

1) Team name
2) Names & emails (cc) of all team members
3) Relevant skills & experience of team members, with link to Github if available. & member role(s) 2/n
4) Brief description of the specific pre-qualifying exercise you intend to submit.

A pre-qualification exercise is due by Jan 20. Based on that, qualifying teams will be selected and announced by Jan 22, 2020. 3/n
Read 6 tweets
The German Army (Amt für Heeresentwicklung, to be exact) has surprised many and published a position paper on #AI for land-based forces. Will translate bits and comment.
Alright, let's do this!
The paper is available here 👇
It was published only some ten days ago, but was finished in August/September.
#AI #KI #Bundeswehr (1/)
augengeradeaus.net/wp-content/upl…
The first thing to note is that with regard to process, this paper is a bit weird: Germany has a national #AI strategy (which mentions military AI in *one sentence*) but no military #AI strategy for the whole of the Bundeswehr. However, now there is this paper for the Army.
(2/)
Read 33 tweets
1/ "Wait, what?" tweetstorm
2/ Wow, #ML powered upsampling has gotten very good. Example video, VGA input, 4K output, from 1990s anime called Yu Yu Hakusho:
3/ To compare, this is a sample of the original, though not the exact one used as input:
Read 6 tweets
We've evolved and here is a #thread of that journey!

Like every successful living species, to survive and get better, you have to evolve. & a company is not different. With the big tech eating into almost every industry it becomes paramount important that traditional businesses
have to think ground up with digital technologies at the core. And today, I'm here to announce that, as a business leader of your company, you don't have to do it alone - Heptagon is there to guide you through your entire digital transformation journey.
We've been creating digitally transformative solutions since 2009. We had an opportunity to work with the NYC government, BMW and we've also built platforms to distribute these solutions. These are our success stories but, we've had our fair share of failures
Read 11 tweets
Alors les amis, The French Ministry of Armed Forces #AI task force has released a new report covering #AI in #defence & a roadmap toward more military adoption. You can read it en français here: defense.gouv.fr/salle-de-press…

Or read on for a few findings & preliminary thoughts.
FR wants its very own JAIC-like cell, the CCIAD, embedded within their Defence Innovation Agency (IAD). (Shout out: FR speakers should follow @KrajeckiMichael). The Coordination Cell for AI in Defence will facilitate efforts on coordination & strategy. p21 for this buried lede
Nice to see that France does not see AI as an end unto itself. Intro is pragmatic & focusses on the weaknesses of second-wave AI systems (although without using the #DARPA-esque vocab), including by noting that #AI does not change the nature of war.
Read 18 tweets
Automation Tools To Ease Your Data Science Project

Working on Data Science Project can be overwhelming for beginners esp student that intend to carry our a research in this field.

Below are some automation tools that data science professional can use.
#DataScience #Thread
1 WEKA
Weka is a collection of machine learning algorithms for data mining tasks. It contains tools for data preparation, classification, regression, clustering, association rules mining, and visualization
#WEKA help you discover practical data mining &learn to mine your own data
#WEKA make it easy for you to play around with data and it gives you test options regarding how you split your #dataset.
I once used WEKA to get result 'clue' for Classification of EEG signal using AIRS, Immuno, CLONALG.

Download WEKA: cs.waikato.ac.nz/ml/weka/
Read 11 tweets
Love the simplicity of #Serverless #FaaS (Function-as-a-Service) but hate the setup process? Look to these 7 open source projects to ease #AWS Lambda deployments: dy.si/M3VJd
———
#BigData #StreamingAnalytics #Cloud #DataScience #MachineLearning #IFTTT #EventDriven #AI
13 free tools for #API design, development, & testing — for example, Amazon API Gateway allows you to build front-end APIs for applications built on Amazon EC2, #AWS Lambda, or any web application:
dy.si/1VTf942
————
#microservices #cloud #serverless #FaaS #coding #IoT
#IFTTT alternatives for developers of #EventDriven workflows: dy.si/ZorG2N
————
#IoT #EdgeAnalytics #EdgeComputing #Microservices #DataScience #BigData #FaaS #ML
————
+See the book “AWS Lambda in Action: Event-driven #Serverless Applications” at amzn.to/2YFDSO8
Read 4 tweets
Department of Medicine Grand Rounds - listening to the amazing patient experience transformation journey led by Babar Hasan and his stellar team (including Muneera Rasheed, Zahra Hoodbhoy, @mominkazi) @AKUGlobal
#mentorship a key component of delivering #compassionate care by nursing - Noreen Sultan @AKUGlobal
#patient experience #transformation using predictive modeling based on #ML - leading with #data @Ephlux @AKUGlobal
Read 3 tweets
A "worrying analysis":

"18 [#deeplearning] algorithms ... presented at top-level research conferences ... Only 7 of them could be reproduced w/ reasonable effort ... 6 of them can often be outperformed w/ comparably simple heuristic methods."

Paper:
lnkd.in/dTaGCTv

#AI
[Updates worth tweeting]

2/
There is much concern about #reproducibility issues and flawed scientific practices in the #ML community in particular & #academia in general.

Both the issues and the concerns are not new.

Isn't it time to put an end to them?
3/
There are several works that have exposed these and similar problems along the years.

👏👏 again to @Maurizio_fd et al. for sharing their paper and addressing #DL algorithms for recommended systems (1st tweet from this thread).

But there is more, unfortunately:
Read 18 tweets
Remember the #BachDoodle? We’re excited to release paper on Behind-the-Scenes design, #ML, scaling it up, and dataset of 21.6M melodies from around the world!
📜 arxiv.org/abs/1907.06637
w/ @fjord41 @ada_rob @notwaldorf @bengiswex Leon Hong @jaxcooo
tl; dr
1/ In three days, people spent 350 years worth of time playing with the Bach Doodle, and the “harmonize” button was clicked more than 55 million times.
2/ The model Coconet 🥥 is an instance of OrderlessNADE and uses Gibbs sampling to generate the harmonizations through rewriting.
📜 Previous blogpost: g.co/magenta/coconet
📝 Paper from #ISMIR 2017: arxiv.org/abs/1903.07227
Read 9 tweets
Recently, the National Crime Records Bureau of India published a tender for the Automated Facial Recognition System (AFRS). Some background (B), and key takeaways (T). #AI #FacialRecognition #ML #India 1/n
B1. The AFRS is conceptualised to modernize the police force, for “criminal identification, verification and its dissemination among various police organizations and units across the country.” 2/n
B2. The rationale: face recognition will help with “facilitating easy recording, analysis, retrieval and sharing of Information” and help in identification of: criminals, missing children/persons, unidentified dead bodies and unknown traced children/persons. 3/n
Read 17 tweets
I’ve been getting a lot of new followers lately, so how about an #epitwitter #ScholarSunday so newcomers can get connected to our lovely community!

Here’s a thread of some of my suggestions for who to follow & why! Add your own in the comments👇🏼
Looking for some academic career advice? Have questions about writing grants? Want tips on making delicious pizza or pie? Head on over to @BillMiller_Epi!
Want support for how to #liveyourbestlife, learn to say ‘no’, or throw an amazing party? You’ll be wanting to follow @lisabodnar!
Read 6 tweets

Related hashtags

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!