AI - Applied Use Cases


WEEKLY REVIEW


Trends


Top hashtags

Top influencers

Miles_Brundage
GoAbiAryan
johnchavens
math_rachel
BulletinAtomic
DeepMindAI
Raspberry_Pi
goodfellow_ian
jackclarkSF

Top sources

fast.ai
ai.facebook.com
cezanneondemand.intervieweb.it
deepmind.com
ruder.io
sites.google.com
stanojevic.github.io
thebulletin.org
automl.org

Get updates live! Follow us on

Choose your newsletters

News

What do machine learning practitioners actually do?

On Jun 14, 2019
@math_rachel shared
RT @ledell: Delighted that @math_rachel from @fastdotai is giving a keynote at the #ICML2019 AutoML workshop! Check out her 3-part blog series about #AutoML: https://t.co/hhKoPE5Avz https://t.co/qwfcEUeVix
Open

It will address what it is that machine learning practitioners do, with Part 2 explaining AutoML and neural architecture search (which several high profile figures have suggested will be key to decreasing the need for data scientists) and Part 3 will cover Google’s heavily hyped AutoML ...

www.fast.ai
On Jun 14, 2019
@math_rachel shared
RT @ledell: Delighted that @math_rachel from @fastdotai is giving a keynote at the #ICML2019 AutoML workshop! Check out her 3-part blog series about #AutoML: https://t.co/hhKoPE5Avz https://t.co/qwfcEUeVix
Open

What do machine learning practitioners actually do?

It will address what it is that machine learning practitioners do, with Part 2 explaining AutoML and neural architecture search (which several high profile figures have suggested will be ...



On Jun 9, 2019
@NathanBenaich shared
RT @graphcoreai: Come and see us at ICML 2019 next week! Find out more about the @graphcoreai Research team. Did we mention we're hiring? https://t.co/RR0dSHCJxi https://t.co/3PTFj0BAug
Open
On Jun 14, 2019
@math_rachel shared
Jeff Dean discusses the importance of computational efficiency: - predicting when to stop unpromising experiments - better training regimens - predicting full dataset accuracy from curacy on a small subset and a shout-out to https://t.co/ktYtgBpxGr progressive resizing :-) https://t.co/fC3CtdGlHc
Open

Making neural nets uncool again

It’s a small world indeed… The dataset collection The following categories are currently included in the collection: Image classification, with a focus on fine-grained classification and ...

On Jun 9, 2019
@Jaguilam2003 shared
RT @lindagrass0: With Artificial Intelligence Fueled by Intel Technology, We will See Unprecedented Developments in Medicine, Scientific Discovery and Education. Link >> https://t.co/UBk9gjEU9P @antgrasso @bkrunner @intel @IntelAI @Inteliot via @lindagrass0 #AI #Tech #DigitalTransformation https://t.co/1wwGYfA8D6
Open

The Intelligence Revolution – Intel’s AI Commitments to Deliver a Better World

With artificial intelligence fueled by Intel technology, we will see unprecedented developments in medicine, scientific discovery and education.

On Jun 13, 2019
@arstechnica shared
I Am Mother envisions a dark future world where AI is almost human https://t.co/YRLHo2xKnI by @JenLucPiquant
Open

I Am Mother envisions a dark future world where AI is almost human

Ars chats with Director Grant Sputore about bringing his first feature film to life.

On Jun 13, 2019
@dmonett shared
RT @jeremyphoward: A curated list of decision, classification and regression tree research papers from the last 30 years with implementations. It covers NeurIPS, ICML, ICLR, KDD, ICDM, CIKM, AAAI etc. https://t.co/JDj6bkTYth
Open

Awesome Decision Tree Research Papers

A collection of research papers on decision, classification and regression trees with implementations. - benedekrozemberczki/awesome-decision-tree-papers

On Jun 11, 2019
@Miles_Brundage shared
RT @GhorbaniAmirata: Tomorrow at #ICML2019 conference, I will be talking about our recent work. We developed a method for equitable valuation of data in #machinelearning called "Data Shapley" https://t.co/ij3WF8cR1P. https://t.co/uDOFuovNk3
Open

Click here to read the article

5/23 Spam Classification Naïve Bayes Classifier 20% mislabeled Flower Classification Retraining Inception-V3 top layer 10% mislabeled T-Shirt/Top vs Shirt Classification ConvNet Classifier ...

On Jun 12, 2019
@johnchavens shared
RT @JessicaFjeld: Have you been following the tide of #AIEthics principles from government, industry, and civil society around the 🌎? We @BKCHarvard have! Pleased to share a draft #DataViz for your feedback: https://t.co/I1GxSavsmY
Open

Introducing the Principled Artificial Intelligence Project

Alongside the rapid development of artificial intelligence, we’ve seen a proliferation of AI “principles,” or guidelines for how AI should be built and used. Governments, companies, ...

On Jun 14, 2019
@DeepMindAI shared
And read more about our prior work on AI for climate change: https://t.co/0uUA4MN2gC
Open

Machine learning can boost the value of wind energy

Since last year, a joint DeepMind/Google project to apply machine learning to 700 MW of wind power in the central US has so far boosted the value of wind energy by ~20%. A neural net ...

On Jun 13, 2019
@olivercameron shared
RT @ali01: While I was at Google X, I worked on an effort to build a robot hivemind––a network of robots with collective intelligence. Our work was part of a broader strategy to gain a data advantage. Here’s what I learned about data network effects. A mega thread 👇 https://t.co/VhKe316vo1
Open

Robot Hiveminds & Network Effects

While I was at Google X, I worked on an effort to build a robot hivemind––a network of robots with collective intelligence. Our work was part of a broader strategy to gain a data advantage. ...

Browse topics

Get updates live! Follow us on

Choose your newsletters

On Jun 12, 2019
@GoAbiAryan shared
Interesting Implementation: 1000x Faster Data Augmentation. PBA is an algorithm that quickly and efficiently learns augmenting data but uses one thousand times less compute, enabling researchers and practitioners to effectively learn using a single GPU. https://t.co/Q9eEjUTnWg
Open
On Jun 11, 2019
@MIT_CSAIL shared
New study shows that object-detection systems are less accurate in lower-income households and on continents like Africa and Asia. https://t.co/nt672aJIHA (v/@facebookai) #ML #machinelearning #BiasInAI https://t.co/l8agwhZQhd
Open

Does object recognition work for everyone? A new method to assess bias in CV systems

Facebook AI researchers have published the first systematic study that measures the accuracy of object-recognition systems for different communities across the world.

On Jun 11, 2019
@Miles_Brundage shared
RT @luisa_zintgraf: I'm very excited to present our work "CAVIA: Fast Context Adaptation via Meta-Learning" at @icmlconf this week (Wed, poster 252) - with @KyriacosShiarli @y0b1byte @katjahofmann @shimon8282 Paper: https://t.co/T51fLBsbSU Blog: https://t.co/LUsLVTXXmE Code: https://t.co/78GKa86jLF https://t.co/RKi4O9IeoP
Open

CAVIA: Fast Context Adaptation via Meta-Learning

[latexpage] Adapting to previously unseen tasks is a long-standing problem in machine learning. Ideally, we

On Jun 9, 2019
@Miles_Brundage shared
RT @ashk4n: The @houseintelcomm is holding a hearing on #AI/#deepfakes this Thur 5/13 @ 9am ET: The National Security Challenge of Artificial Intelligence, Manipulated Media, and “Deepfakes" Witnesses are: @daniellecitron @jackclarksf and David Doermann (UMD) https://t.co/svjbedToTD
Open

House Intelligence Committee To Hold Open Hearing on Deepfakes and AI

Among other topics, the Committee will seek testimony on: Future advancements in deepfake technology; Detecting and tracking deepfakes; How deepfakes could allow people to deny legitimate ...

On Jun 11, 2019
@MelanieHannah shared
RT @lizpelly: Spotify knows its most valuable asset to multinational corporations: its troves of data about our emotions. My latest for @thebafflermag https://t.co/cTNXYShg1p
Open

Big Mood Machine

Inside Spotify’s emotional surveillance-driven quest for total ad domination.

On Jun 9, 2019
@GoAbiAryan shared
RT @ch402: For the last few years, one of the staples of my research has been visualizing individual neurons in vision models. But that's only a partial picture -- neurons work together. Activation Atlases are a way to explore the space neurons jointly represent. https://t.co/TEa9WuYBMg https://t.co/yb3zvpxmhq
Open

Exploring Neural Networks with Activation Atlases

By using feature inversion to visualize millions of activations from an image classification network, we create an explorable activation atlas of features the network has learned and what ...

On Jun 13, 2019
@max_read shared
turns out AI is actually an omnicidal threat to the planet, just not in the way the hucksters have been claiming https://t.co/lxhUdQkDsa
Open