FOSS Machine Learning News week 23-2020

Welcome to our biweekly selection of Free and Open machine learning news. Created using our own opinionated selection and summary algorithm. FOSS machine learning is crucial for everyone. Machine Learning is a complex technology. So keep it simple.

1 Eye-catching advances in some AI fields are not real

Fraud. I do not like it. Great research which one again shows how important openness is!


2 The Seven Tools of Causal Inference, with Reflections on Machine Learning

I really like the amazing picture in this article. A bit like Escher. Most important message from this article: Data science is a two-body problem, connecting data and reality, including the forces behind the data. Something you will forget when working behind notebooks and algorithms.


3 Undergraduates develop next-generation intelligence tools

In a collaboration with Massachusetts General Hospital, MIT Quest is evaluating whether automation could help to boost the nation’s supply of viable livers. The project involves training a deep neural network to pick out globules of fat on liver tissue slides to estimate the liver’s overall fat content. His goal is to teach the robots to pair sights with sounds, and to exploit this information to build better representations of their environment. She then ran a set of online experiments to see how human subjects would interpret analogous sentences in a story. He’s working with Pulkit Agrawal, an assistant professor in MIT’s Department of Electrical Engineering and Computer Science, who is interested in designing algorithms with human-like curiosity.

(MIT Reseach CS)

4 Ear2Face: Deep Biometric Modality Mapping

Great research on how an ear can be used to create a face. Of course creepy technology we normally see only in Hollywood films. Unfortunate the code is not yet released, so playing with this code and maybe your own ear has to wait a bit longer.


5 A tool to detect automatically generated text

I love complex security challenges. Nowadays figuring out if the text you read is created by a human or by an algorithm is a real brain breaker. This article and code helps solving this problem. And the demo page is slow but great.


6 Jukebox

A neural net that generates music, including rudimentary singing. The code is available for us all to give it a try! If you are short on time you can also launch the notebook to see how it is done. The created songs are worth to take a quick look. As a promoter of open & safe ML, I like the lyrics of the Simon and Garfunkel attempt.


7 A European way towards sustainable AI

A European way towards sustainable AI. Nice article with some quotes to think about. Of course I rather see that we all promote key open principles for technology innovation. And all make sure we keep ML/AI software free from patents. The patents system is broken and should not be repaired…


8 Acme – A framework for distributed reinforcement learning

I like frameworks that try to make complex things simpler. The explanation with visuals in the docs makes the impression that Acme is indeed simple to use. Acme strives to expose simple, efficient, and readable agent baselines while still providing enough flexibility to create novel implementations. This ML framework will be incorporated in an update of the FOSS ML Guide.


9 Zero-Shot Learning in Modern NLP

State-of-the-art NLP models for text classification without annotated data. Great blog on how to use state-of-the-art NLP models for sequence classification without large annotated training sets.


10 Go AI Past, Present, and Future

There’s something magical about the game of Go. For thousands of years, it has captured the imagination of those who want to learn what it is to learn, to think about what thinking means. AI can beat top professionals now. This blog article is tracing the history of the game, why it remained so difficult to beat humans for so long, and what the future of Go may hold.


The FOSS Machine Learning News Blog is a brief overview of open machine learning news from all over the world. Free and Open machine learning means that everyone must be able to develop, test and play and deploy machine learning solutions. Read and share the FOSS ML Guide! And remember:You are invited to join the Free and Open Machine Learning open collaboration project.