Top updates related to Artificial Intelligence during week 49 of 2020, weekly edition of This week in AI
A summary of recent updates in the field of Artificial Intelligence (AI) for week 49 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 49 of 2020
This section is for recent industry updates related to Artificial Intelligence.
Google’s astonishing revelation about ML: The phenomenon of “Underspecification” is considered to be a potential limitation in ML applications. To know more, read here.
MONAI: An open-source framework for medical imaging and healthcare using AI is launched. Here is its entire coverage for a glance.
Linformer by Facebook: Facebook’s new hate speech and violence content analyzer are based on transformer learning called “Linformer”. The algorithm has successfully removed 94.7% of content that is slanged or misspelled. To get more information, visit here.
AI for Cancer cell elimination: AI can now help develop doctors treat cancer cells through the suggestion of drug combination. This will reduce the diagnosis time and improve life expectancy. Here is the entire article for a quick read.
Phantom Menace: EpiSci developed an autonomous fighter agent giving a fighter pilot a true to life virtual experience mid-air. To know more, visit here.
AI for placing Loon’s Baloon: High-flying balloons are helping remote nations by providing them broadband connections but keep them in place is also another task on its own. Google’s Loon has been used to predict wind speed and direct the balloon to the right position. Here is the full coverage for reference.
Waymo’s city for testing cars: Waymo is due to open two facilities for testing its driverless cars including a playground. Such an environment will surely help fine-tune the performance and decision making of autonomous vehicles in dangerous test scenarios. To know more, visit here.
NVIDIA’s Neural Talking synthesis: The model is called a talking head synthesis tool that intends to improve bandwidth and video efficiency of video conferencing tools. Get a glance at the story over here.
Articles for week 49 of 2020
This section is for recent Blog Posts/Tutorials related to Artificial Intelligence.
DeepMind solves protein structures problem: DeepMind has been successful in determining the 3D shape of proteins from amino structures. Their research is published in Nature’s research series and can be found here.
One GPU with low training time: This is a blog post explaining fast and stable GAN that involves lightweight parameters thus reducing the training time. The layer configuration changes have led to an increased speed.
Transfer learning for segmentation: This is a blog explaining transfer learning for road crack detection with as low as 100 images. Here is the full article for a quick read.
DL journey update: The blog post covers the writer’s experience in starting with Neural Style learning. A pretty handful yet must-try projects explained and executed. Read the article here.
This section is for recent Tools/Applications related to Artificial Intelligence.
This section is for recent Lectures/Conferences/Webinars related to Artificial Intelligence.
In conversation with Google-Brain Researcher: This is an informative conversation with Sara Hooker who is an ML researcher at Google-Brain. She conveys her thoughts on interpretability, some state-of-the techniques for every Ml engineer. Here is the Youtube video link.
Auto-encoders: An intuitive video for understanding the working of Auto-encoders and representation learning. Also, its two prime applications of data generation and denoising are discussed.
Clinical NLP: This video gives a brief description of clinical Natural Language processing along with its challenges, tasks involved and datasets available.
This section is for recent research published related to Artificial Intelligence.
Benchmark for cementing left atrium: A research published by various creative minds in developing benchmark results for segmenting left atrium from LGE-MRI and CNN.
Graph Neural Networks for ENSO forecasting: The Graph Neural Networks have been found to outperform the CNN based deep learning algorithms for seasonal forecasting such as ENSO. To get the full article, visit arXiv here.
ML-based flood forecasting: The intimation for floods would save a lot of lives along with the economy experiencing it. This article reviews two major components playing role in flood forecasting and can be accessed here.
Analyzing memes on Reddit: The research is based on content-based prediction analysis that finds the reason behind certain memes going viral. The algorithm uses image and text analysis both for prediction and can be read here.
This section is for updates for other various related topics to Artificial Intelligence.
Lab-ML App: This a handy application to track the progress go your models being trained in the form of push notifications. The installer is a one-line command and it is ready to use. Try it out here.
That’s all for the week 49 of 2020. In case you think we have missed out on 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.