A summary of recent updates in the field of Artificial Intelligence (AI) for week 41 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 41 of 2020
This section is for recent industry updates related to Artificial Intelligence.
- COVID-19 tracking with the help of Google: A teaming of CMU and Google to build COVIDcast that will monitor, forecast, and help public health services to make appropriate decisions is gaining pace. The project is led by Ryan Tibshirani, a faculty member at CMU.
- Identity protection using deep fakes: In a documentary produced by HBO used deep fakes to protect the identities of the LQBTQ interviews that they conducted. Here is the entire post.
- PwC-Powered Code Tab: The tie-up of PwC and arXiv has led to a new feature on their landing page for ML articles. This tab would help the ML community to access the latest research.
- Waymo’s Driverless rides in Arizona: After test driving by few hundred riders in 2019, Waymo have now made fully driverless rides to all its users in the Phoenix area. Here is the entire story.
- Farewell to CNN and Welcome Transformers: The ML community has welcomed a research work submitted to the ICMR 2021 with open hands due to its eye-catching contribution in Image recognition. The technique aims to outperform the state-of-the-art CNNs with the transformers. Read the full story here.
- Middle East to be leading the AI hub: The world’s first AI university MBZUAI in UAE has launched its AI program in 2019. A bunch of AI research problems being currently addressed in the UAE will contribute to billions of USD soon. Here is the entire article.
- Hard to build ethical ML framework for Healthcare: A descriptive write-up on reasons behind the hardships faced by many researchers in building ethical ML frameworks. Right from overcoming bias data in problem selection to the aftereffects of deployment, are all discussed in this article.
- New age Farming by Robot reapers: Due to increased labor shortage and climatic imbalances, John Deere is helping farmers to grow more through less effort and damage to the land. They are certainly using AI and ML to extend their support in all possible ways they can. To know more, visit here.
- Why AI can’t reach its fullest without physical medium: A philosophical perspective of developing AI an expanding it in all directions while experiencing a setback of disembodiment. The entire article can be accessed here.
- Oxygen cylinder prediction for COVID-19: NVIDIA along with the Mass General Brigham Hospital has developed an algorithm that can predict the requirement of oxygen cylinders amongst the community. The analysis was based on data gathered from 20 Hospitals in 20 days. Read the entire post here.
- Xbox series X offers superb results through auto-HDR: The Xbox X series is using ML to auto-adjust through High Dynamic Range that enhances resolution making older flat panels look even better. Here is the full post.
- Top AI Researchers: With AI being explored in all domains, the top AI figures who influence the society through re-defining applications of Deep Leaning is aired by Analytics India Magazine. The complete list can be accessed here.
- Top AI tech giants: A nice read as Big Nine Titans or companies that could lead to a greater loss to humanity if they do not flourish. Here is the whole book review.
- AI-powered Education: AI has changed the shopping experience and entertainment is personalized on a different scale. The use of annotations in text, audio, and video analysis has helped the e-learning community get better recommendations. The pandemic of course was not foreseen, but e-learning had already begun to gain momentum bring knowledge to all. The article here mentions AI plays a key role in modeling e-learning platforms and contributing monetarily as well as mentally.
- Google Nest Review: An honest review of Google Nest is offering good sound at a price less than the Google Home which was launched back in 2016. Here is the link if you need a detailed review.
- RAISE 2020 announcements at a glimpse: The Responsible AI for Social Empowerment Summit was held this week and was organized by MeitY. The summit aimed to drive India towards a data-driven society. The top announcements made at this summit can be viewed here.
Articles for week 41 of 2020
This section is for recent Blog Posts/Tutorials related to Artificial Intelligence.
- Debugging Tensorflow: An efficient time saving debugging checklist for Tensorflow that enhances code development accuracy is explained here. Right from types of bugs to saving checkpoints to contact tracing precisely elaborated in this article.
- Handling Pandas Dataframe: This is a comprehensive article in performing functions on the pandas Dataframe for Data wrangling. To know more, read this.
- Scaling Multi-agent RL using Game Theory: This article is based on pointing out the challenges of Centralized learning and proposes Diff DAC learning MARL (Multi-Agent Reinforcement Learning) scenarios. To know more, visit here.
- AI sailing upwind: A simulation experience of Sailing through Machine Learning can be found here. The technique used here is Q-learning to solve two different tasks of sailing in the sea and channel.
- NVIDIA NeMo: Neural Modules and Models (NeMo) for Conversational AI is a NVIDIA’s open-source toolkit for conversational AI built using PyTorch. It is a multi-application kit meant for Speech Recognition, NLP, and Speech Synthesis.
This section is for recent Tools/Applications related to Artificial Intelligence.
This section is for recent Lectures/Conferences/Webinars related to Artificial Intelligence.
- Unsupervised Multi-Document Summarization: This video is a walkthrough of a research article on Multi-document summarization using the Neural Document Model and is said to outperform existing unsupervised learning techniques.
This section is for recent research published related to Artificial Intelligence.
- Deformable DETR: The DETR model is known to automate the object detection task, but suffers from slow convergence, spacial issues, and transformer attention. The Deformable DETR overcomes these shortcomings and uses attention modules to work on a small set of sampling points. Read the full paper here.
This section is for updates for other various related topics to Artificial Intelligence.
That’s all for the week 41 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.