A summary of recent happenings in the field of Artificial Intelligence (AI) for week 37 of 2020. For more regular updates and related discussions, join us in the Artificial Intelligence Community. As always, thanks for your support for sharing it with your friends and colleagues on social media.

Industry updates for week 37 of 2020

This section is for recent industry updates related to Artificial Intelligence.

  • Business value of ML: A book called Prediction Machines explains the economic perspective of ML and how it will and has influenced various economies. To read the entire article visit here.
  • Our Universe is made of Neural Networks? : A physics professor at the University of Minnesota Duluth has attempted to define the entire universe as a Neural Network in his paper submitted to arXiv. To read more about this, visit here.
  • Google’s attempt to play safe: Just weeks before US presidential elections, Google has fine-tuned its ‘Autocomplete’ feature to avoid any controversies or involvement in elections. A well-played move by Google, to read more details, visit here
  • An Award-winning face recognition tool: Recently, after some western countries facial recognition software false classified a black African-American civilian, a new tool based on black people dataset has won an Africa Prize for Engineering Innovation. A research lead by Charlette N’Guessan used BASE API and AI to verify people’s identity using their cell phones. To get a detailed read, visit here.
  • AI and its moral risk: AI involves taking decisions and assuming certain Unknowns. As easy as it sounds, everything comes with a price, assuming the threat to humanity through this evolving technology that we have created. This piece of the article touches on some aspects of AI and its risks involved.
  • US Air Force has Dog robots: The best fictional scene is now a reality through the dog robots which were deployed to the airstrip to check potential human threats. It is a part of Advanced Battle Management Systems (ABMS). To read more about what happened on that airstrip, read here.
  • Robot writes human-like article: This newsreel is going viral and inducing concerns over AI abilities to pose a human threat. OpenAI’s GPT3 generated an article worth reading and with high creative skills. 
  • Transformers are Graph Neural Networks: Apart from recommendation systems, the Graph Neural Networks might be useful in the NLP field as well. To read on representation learning, visit here.

Articles for week 37 of 2020

This section is for recent Blog Posts/Tutorials related to Artificial Intelligence.

  • Choosing the right GPU: A nice blog on various GPU’s based on the author’s experience is a must-read for a newbie as well as a professional. Right from, the memory, speed, and types to which GPU’s to avoid are all on the list of things explained in the post. 
  • Pseudo-LiDAR: This is an article on Stereo-vision for self-driving cars. The Light Detection and Range (LiDAR) sensors that return depth information along with certain geometrical skills are a cost-saving approach.
  • Project Zeus: A beginner’s guide for detecting rooftops is effectively explained in this article. The feature type, training algorithm, and results are also shared in this post.
  • Proximal Policy Optimization(PPO) with Tensorflow: A reinforcement learning algorithm, PPO that is developed on swift for Tensorflow This google summer project is available for download here.
  • The AI hoax: The article wants to awaken researchers about overboosting of certain AI solutions which are claimed to be geniuses but in fact may be bad choices. Because with changing scenarios and trails, the algorithms respond vary and no one solution is complete on its own. 
  • Color Recognizer in Python: This is a simple color detection project that is based on Pandas and OpenCV. To know the complete steps, visit here
    .

Softwares

This section is for recent Tools/Applications related to Artificial Intelligence.

  • Deep HDR imaging: Constructing a high dynamic range (HDR) images from multiple low dimensional images is challenging. This new approach that employs a novel deep learning framework called NHDRRNet removes ghosting artifacts. The Github script is available to play around.
  • Create Cartoons using Tensorflow lite: This is a complete tutorial Tensorflow to cartoonize an image using Tensorflow Lite and further create an application of it. The blog can be explored here.
  • PyGLN, Gated Linear Network: An implementation of Gated Linear Networks using Numpy, PyTorch, Tensorflow, and JAX is based on a recent publication by DeepMind. The Github code is available here.
  • Sentiment Sam: A must-try demo of Sentiment Sam will show you the fun side of AI. It is a robotic poet who will sing you a song. Mind it, be patient.
  • BLM photo Anonymization: A working project developed by Stanford ML team and LSC-CNN covers photos of protestors with emoji circles. Here is a demo, if you wish to try.

Videos

This section is for recent Lectures/Conferences/Webinars related to Artificial Intelligence.

  • Top Conferences in Deep Learning: Those of you who wish to attend, submit, present their research in top deep learning conferences, here is a list to follow. 
  • Robots replacing Tow-truck: A wonderful video portraying spot robots pulling a truck is a creation by New Boston Dynamics. This seems to give a sneak-peak into the future of automation. 
  • Humanoid robot Aida’s drawings on display: This human-like robot is great at drawing real-life portraits and is on show in the UK. The video of Aida at work is available here.
  • Jupyter Notebook: A necessary prerequisite for beginners in the field of ML is Jupyter Notebook. The video shows an overview of handling it and some other significant features.
  • Tensorflow Basics: This video is part of Tensorflow Beginner’s series on Binary Classification. 
  • Introduction to ML-K nearest neighbors: A part of the lecture series by Sebastian Raschka on K nearest neighbors is a superb video for understanding its scikit-learn implementation.
  • Work from Phone: Work from home and work from phone are trending during these Pandemic times. This video shows how to complete tasks on your phone.

Research

This section is for recent research published related to Artificial Intelligence.

  • Generative Language Modeling: The OpenAI’s research on the application of the Transformer based Language model for Automated Theorem Proving. The article is available here

Miscellaneous

This section is for updates for other various related topics to Artificial Intelligence.

  • Machine Learning Engineering: Andriy Burkov’s e-book on Machine Learning has been released and can be purchased here.

That’s all for the week 37 of 2020. In case you think we have missed out something and that is significant to add, please free to reach out and submit to us via our Contact form or via Social media channels: Twitter and Facebook.