Artificial Intelligence Weekly – Artificial Intelligence News #83

Guide to Game AI, AI for cybersecurity, Hiring in data science teams, Facebook ML video series and more
 

Artificial Intelligence Weekly

 

ARTIFICIAL INTELLIGENCE NEWS #83  August 16th 2018

 

  In the News  

 

AI for cybersecurity is a hot new thing—and a dangerous gamble

Machine learning and artificial intelligence can help guard against cyberattacks, but hackers can foil security algorithms by targeting the data they train on and the warning flags they look for.

technologyreview.com

 

Face Off: Law enforcement use of face recognition technology

“Face recognition is poised to become one of the most pervasive surveillance technologies, and law enforcement’s use of it is increasing rapidly. Today, law enforcement officers can use mobile devices to capture face recognition-ready photographs of people they stop on the street; surveillance cameras boast real-time face scanning and identification capabilities; and federal, state, and local law enforcement agencies have access to hundreds of millions of images of faces of law-abiding Americans. On the horizon, law enforcement would like to use face recognition with body-worn cameras, to identify people in the dark, to match a person to a police sketch, or even to construct an image of a person’s face from a small sample of their DNA.”

eff.org

 

Data-mining medieval text reveals medically bioactive ingredients

Medieval apothecaries used recipes with significant antibacterial properties, researchers say.

technologyreview.com

 

 

  Learning  

 

The total beginner’s guide to Game AI

This article will introduce you to a range of introductory concepts used in artificial intelligence for games (or ‘Game AI’ for short) so that you can understand what tools are available for approaching your AI problems, how they work together, and how you might start to implement them in the language or engine of your choice.

gamedev.net

 

Facebook Field Guide to Machine Learning video series

Six-part video series developed by the Facebook ads machine learning team. The series shares best real-world practices and provides practical tips about how to apply machine-learning capabilities to real-world problems.

fb.com

 

Running Agile Machine Learning Experiments

medium.com

 

Learning Math for Machine Learning

ycombinator.com

 

 

  Sponsor  

 

Build Modern Chatbots Fast

Create modern conversational chatbot experiences in your web, desktop, and mobile applications with the Conversational UI components from Progress Telerik and Kendo UI. These Chatbot framework agnostic tools enable .NET and JavaScript developers to build high-performant, modern chatbots with an outstanding UI. Get your 30-day free trial today.

telerik.com

 

 

  Software tools & code  

 

9 things you should know about TensorFlow

hackernoon.com

 

Classification datasets results

Summary of the current state of the art in image classification.

github.io

 

 

  Workplace  

 

Hiring for autonomy in Data Science teams

Interview with Eric Colson, Chief Algorithms Officer for Stitch Fix, an online subcription personal shopping service, on Stitch Fix’s engineering culture works, including what they value, what they look for in new hires, and how they’ve architected their platform to enable astounding success.

soundcloud.com

 

Are you hiring?

I’d be happy to include your job listing in AIWeekly. (just so you know, AIWeekly currently has ~14000 subscribers)

Fill out this form to add an open position to the job board. Thank you!.

 

 

  About  

 

This newsletter is a collection of AI news and resources curated by @dlissmyr. If you find it worthwhile, please forward to your friends and colleagues, or share on your favorite network!

Share on Twitter · Share on Linkedin · Share on Google+

Suggestions or comments are more than welcome, just reply to this email.

Thanks!

 

You received this email because you subscribed via the Artificial Intelligence Weekly site.
We’ll be sorry to see you go but you can unsubscribe instantly.
 
Kozelo Boite 37 16 bis rue d’Odessa 75014 Paris France
 
 

 
 
©2018 Artificial Intelligence Weekly | Privacy Policy | Image moderation
 
Published with Curated

Artificial Intelligence Weekly – Artificial Intelligence News #82

The media doesn't get AI, program synthesis, evolutionary algorithms, Auto ML, Edge TPU and more
 

Artificial Intelligence Weekly

 

ARTIFICIAL INTELLIGENCE NEWS #82  August 2nd 2018

 

  In the News  

 

'The discourse is unhinged': how the media gets AI alarmingly wrong

Social media has allowed self-proclaimed ‘AI influencers’ who do nothing more than paraphrase Elon Musk to cash in on this hype with low-quality pieces. The result is dangerous.

theguardian.com

 

Beijing’s big brother tech needs African faces

"Zimbabwe is signing up for China's surveillance state, but its citizens will pay the price."

foreignpolicy.com

 

Also in the news…

  • ACLU says Amazon's face recognition Falsely matches 28 members of Congress with mugshots. More than anything else, this is a reminder that face recognition through neural networks is probabilistic by nature and must be handled correctly (Amazon says ACLU did not use a proper threshold). More
  • Pentagon said to be preparing major AI strategy this summer. More

 
 

  Learning  

 

Natural Language Processing is Fun!

Introduction to NLP: sentence segmentation, tokenization, speech prediction, lemmatization, stop word identification and dependency parsing.

medium.com

 

Program Synthesis in 2017-18

Great overview of recent advances in Program Generation.

alexpolozov.com

 

Evolutionary algorithm outperforms deep-learning machines at video games

Neural networks have garnered all the headlines, but a much more powerful approach is waiting in the wings.

technologyreview.com

 

An opinionated introduction to AutoML and Neural Architecture Search

AutoML and Neural Architecture Search — where you let the algorithms choose and optimize neural networks' architectures and hyperparameters — are two hot topics right now. Some people like Jeff Dean even claim that they will help replace machine learning expertise with raw computational power.

fast.ai

 
 

  Software tools & code  

 

Measuring abstract reasoning in neural networks

DeepMind introduces an abstract reasoning dataset similar to some tasks found in human IQ tests.

deepmind.com

 

How decision trees work

github.io

 
 

  Hardware  

 

Edge TPU

Google announced a specialized TPU chip for edge devices. Still waiting for details on performance.

techcrunch.com

 
 

  Workplace  

 

Hiring? Looking for a job?

I am listing open positions in A.I. / M.L. here. Feel free to submit open positions here!

 
 

  Some inspiration  

 

The ethics of computer science: this researcher has a controversial proposal

Nature talks to Brent Hecht, who says peer reviewers must ensure that researchers consider negative societal consequences of their work.

nature.com

 

Why is Google Translate spitting out sinister religious prophecies?

Turns out that repeating the word "dog" many times led Google's neural networks to perform fun (some say creepy) translations.

vice.com

 
 

  About  

 

This newsletter is a collection of AI news and resources curated by @dlissmyr. If you find it worthwhile, please forward to your friends and colleagues, or share on your favorite network!

Share on Twitter · Share on Linkedin · Share on Google+

Suggestions or comments are more than welcome, just reply to this email.

Thanks!

 
You received this email because you subscribed via the Artificial Intelligence Weekly site.
We’ll be sorry to see you go but you can unsubscribe instantly.
 
Kozelo Boite 37 16 bis rue d'Odessa 75014 Paris France
 
 

 
 
©2018 Artificial Intelligence Weekly | Privacy Policy | Image moderation
 
Published with Curated

Artificial Intelligence News — Newsletter on Deep Learning & AI – Artificial Intelligence News #81

IN THE NEWS

  • Microsoft asks the U.S. Congress to regulate face recognition. More
  • A report by the Australian Strategic Policy Institute urges better oversight of international partnerships on AI, to ensure that collaborations are not being exploited for military uses. More
  • Facebook hires head of chip development from Google. More

LEARNING



SOFTWARE TOOLS & CODE



WORKPLACE

Hiring? Looking for a job?

I am listing open positions in A.I. / M.L. here. Feel free to submit open positions here!

SOME INSPIRATION

Artificial Intelligence News — Newsletter on Deep Learning & AI – Artificial Intelligence News #80

WELCOME

Are you hiring?

I will soon start listing open positions in the newsletter, and I’d be happy to include your job listing in AIWeekly for free. (just so you know, AIWeekly currently has ~14000 subscribers)

Fill out this form to add an open position to the job board. Thank you!

IN THE NEWS



LEARNING




SOFTWARE TOOLS & CODE



Deep Learning Weekly – 🤖 – Issue #87: AI Principles, 12TB RAM, AI Bubble, Tiny Devices, Software Stack 2.0, One-Shot Object Detection, Create ML and more… | Revue

 

Hey Folks! This week, we learn about Google’s AI principles, Microsoft announced machines with huge a
 
June 14 · Issue #87 · View online
Deep Learning Weekly

Hey Folks!
This week, we learn about Google’s AI principles, Microsoft announced machines with huge amounts of memory, first thoughts about an AI bubble begin to surface and we take a look at machine learning on tiny devices.
Andrej Karpathy shares some interesting views on the software stack 2.0 at Tesla, we learn everything about one-shot object detection and AutoAugment from Google, take a look at Apples new Create ML tool and get to know a machine learning toolchain built for artists.
As we just crossed 8000 subscribers, we would like to thank you once again for all of your support. As always, if you want to help us grow this great community of deep learning enthusiasts, simply share this issue with friends and colleagues.
See you next week!


Industry

AI at Google: our principles


Microsoft Azure will soon offer machines with up to 12 TB of memory


When the bubble bursts…


Why the Future of Machine Learning is Tiny

Why This Startup Created A Deep Learning Chip For Autonomous Vehicles

Sponsored Link

Mention DLWEEKLY for $400 USD discount on any Lambda Quad!
Learning

Building the Software 2.0 Stack by Andrej Karpathy from Tesla


One-shot object detection


Improving Deep Learning Performance with AutoAugment

Libraries & Code

Training a Text Classifier with Create ML and the Natural Language Framework

LSTM · ml5js

Papers & Publications

Improving Language Understanding with Unsupervised Learning

Why do deep convolutional networks generalize so poorly to small image transformations?

Did you enjoy this issue?


If you don’t want these updates anymore, please unsubscribe here
If you were forwarded this newsletter and you like it, you can subscribe here
Powered by Revue
 

 

Deep Learning Weekly – 🤖 – Issue #88: OpenAI Five, Twitter Workflows, Speech Synthesis, Inverse RL, DensePose, Confusing TensorFlow and more… | Revue

 

Hey and welcome to another week in deep learning! This week, we obviously start off with OpenAIs move
 
June 27 · Issue #88 · View online
Deep Learning Weekly

Hey and welcome to another week in deep learning!
This week, we obviously start off with OpenAIs move into Dota 2 gaming, look at ML Workflows at Twitter, learn about reduced GPU pricing on GCP and explore speech synthesis services as well as depth inference for photo editing. We learn about inverse reinforcement learning, object detection, and photo caption. Afterward, we get a great TensorFlow concepts explanation from a Google Brain resident, get to know Facebooks DensePose, a new portal linking papers and code, and the best paper of CVPR2018.
As always, we hope you’ll enjoy reading as much as we did and would appreciate you sharing this newsletter with friends and colleagues.
You may have noticed, that we’re currently running on a bi-weekly schedule due to various reasons, but hope to bring the newsletter back to the normal pace soon. To make up for the delays, we just added a few more links to this weeks issue.
Happy reading and hacking!


Industry

OpenAI Five


Productionizing ML with Workflows at Twitter


Introducing improved pricing for Preemptible GPUs


Speech Synthesis as a Service


From 2D to 3D Photo Editing

Learning

Learning from humans: what is inverse reinforcement learning?


Understanding Deep Learning for Object Detection


How to Develop a Deep Learning Photo Caption Generator from Scratch

Libraries & Code
Tensorflow: The Confusing Parts


Introducing Apex: PyTorch Extension with Tools to Realize the Power of Tensor Cores


Facebook open sources DensePose


NLP-progress: Repository to track the progress in Natural Language Processing (NLP)

Papers & Publications
Papers with Code: The latest in machine learning

Taskonomy: Disentangling Task Transfer Learning

On Calibration of Modern Neural Networks

Did you enjoy this issue?


If you don’t want these updates anymore, please unsubscribe here
If you were forwarded this newsletter and you like it, you can subscribe here
Powered by Revue
 

 

Deep Learning Weekly – Issue #90 – 🤖- Troubling Trends in Machine Learning Research, Design Patterns for Production NLP System & Improving Connectomics | Revue

 

Howdy folks and welcome to another week in deep learning! This week we take a look at the ominous use
 
July 19 · Issue #90 · View online
Deep Learning Weekly

Howdy folks and welcome to another week in deep learning!
This week we take a look at the ominous uses of AI by the Chinese government, some troubling trends in machine learning research, we learn about feature wise transforms and design patterns for production NLP systems
A fun little bit of trivia, the whimsical Not Hotdog App known from the HBO show Silicon Valley was built by fast.ai student and is now nominated for an Emmy.
Happy reading and hacking!


Industry

An Overview of National AI Strategies – Politics + AI


Inside China’s Dystopian Dreams: A.I., Shame and Lots of Cameras


Apple’s New AI Chief Takes on Oversight of Siri

Facebook AI Research Expands With New Academic Collaborations

Troubling Trends in Machine Learning Scholarship

Sponsored Link

Deep Learning Weekly has secured a £200 discount off registration, quote DLW200!
Learning

Design Patterns for Production NLP Systems


Feature-wise Transformations


Using Deep Learning to Automatically Rank Millions of Hotel Images

Libraries & Code

Switchable-Normalization


A Project Based Introduction to TensorFlow.js – Knowledge-Exploration Systems

Papers & Publications
Google AI Blog: Improving Connectomics by an Order of Magnitude


Glow: Better Reversible Generative Models


Proceedings of Machine Learning Research

An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution

Did you enjoy this issue?


If you don’t want these updates anymore, please unsubscribe here
If you were forwarded this newsletter and you like it, you can subscribe here
Powered by Revue
 

 

Deep Learning Weekly – 🤖 – Issue #89: Fairness, Driving in a Day, Fake SloMo, L2 Regularization, RNNs, Art, Prototyping with TF, Edge Performance and more…

 
Hey and Welcome to a new week in deep learning! This time, we take a look at bias detectives ensuring
 
July 10 · Issue #89 · View online
Deep Learning Weekly

Hey and Welcome to a new week in deep learning!
As always, we hope you’ll enjoy reading as much as we did and would appreciate you sharing this newsletter with friends and colleagues.
See you next week!


Industry

Bias Detectives: The Researchers Striving to Make Algorithms Fair


Facial recognition software is not ready for use by law enforcement

Learning to drive in a day


Nvidia uses artificial intelligence to fake realistic slow-motion video


Join Company Builder Entrepreneur First (Sponsored)

Learning
A New Angle on L2 Regularization


Recurrent Neural Networks: The Powerhouse of Language Modeling


My AI-Art Talk for Resonate


Amazon.com sales rank data for Kindle and print books (61,000 books, 200,000,000 data points)

Libraries & Code
Patterns for Fast Prototyping with TensorFlow

Train a model in tf.keras with Colab, and run it in the browser with TensorFlow.js

How fast is my model?

Conferences (Sponsored)

O'Reilly Artificial Intelligence Conference in San Francisco


Living in an AI World: How Startups are Shaping Tomorrow

Papers & Publications

Capture the Flag: the emergence of complex cooperative agents

Self-Supervised Tracking via Video Colorization

Did you enjoy this issue?




If you don't want these updates anymore, please unsubscribe here
If you were forwarded this newsletter and you like it, you can subscribe here
Powered by Revue
 
 

Artificial Intelligence Weekly – Artificial Intelligence News #79

OpenAI beats humans, OpenStreetCam, the state of AI in 2018, why businesses fail at ML, red flags for data science interviews and more
 

Artificial Intelligence Weekly

 

ARTIFICIAL INTELLIGENCE NEWS #79  July 5th 2018

 

  In the News  

 

OpenAI algorithms beat human teams at computer game Dota 2

Some say this is as big a step as AlphaGo beating Lee Sedol 2 years ago. Beating a team of humans at Dota2 is no small feat indeed. It requires coordination and requires handling an extremely complex gameplay with many possible moves, fog, long-term planning etc. See also OpenAI’s blog post.

technologyreview.com

 

Also in the news…

  • AI system beats chinese doctors at brain diagnosis competition. See summary video
  • Google predicting when a patient will die — first steps of a broader push by Google into health and healthcare. More
  • China takes surveillance to new heights with flock of robotic Doves. More

 

 

  Learning  

 

156 slides on the state of AI

stateof.ai

 

2018 Big Data & AI Landscape

Overview of the current state of A.I. by Matt Turck (FirstMark Capital). What’s changed over the past year, where things are going, and the most important players.

mattturck.com

 

 

  Software tools & code  

 

An AI speed test shows clever coders can still beat tech giants like Google and Intel

theverge.com

 

Self-Supervised Tracking via Video Colorization

By teaching an algorithm to correctly colorize videos and maintain consistent colors between frames, you end up building a network that is very good at tracking objects and human positions.

googleblog.com

 

Open Street Cam

An open version of Google Street View. May prove very useful for map-making, navigation, autonomous driving and more.

improve-osm.org

 

 

  Workplace  

 

Why businesses fail at machine learning

Taking a cooking analogy, should businesses focus on building appliances — the underlying ML algorithms / infrastructure — or building cooking recipes using existing ML tools?

hackernoon.com

 

Red flags in Data Science interviews

If you are getting interviewed for a Data Science position, you should take time to evaluate the company you are interviewing with. Here are some red flags to watch out for.

hookedondata.org

 

 

  Some inspiration  

 

Ways to think about machine learning

ben-evans.com

 

You received this email because you subscribed via the Artificial Intelligence Weekly site.
We’ll be sorry to see you go but you can unsubscribe instantly.
 
Kozelo Boite 37 16 bis rue d’Odessa 75014 Paris France
 
 

 
 
©2018 Artificial Intelligence Weekly | Privacy Policy | Image moderation
 
Published with Curated