Week Ending 4.26.2020
RESEARCH WATCH: 4.26.2020
Over the past week, 75 new papers were published in "Computer Science - Artificial Intelligence".
The paper discussed most in the news over the past week was by a team at Google: "AutoML-Zero: Evolving Machine Learning Algorithms From Scratch" by Esteban Real et al (Mar 2020), which was referenced 21 times, including in the article This Is How Algorithms Will Evolve Themselves in Popular Mechanics. The paper also got the most social media traction with 1550 shares. A user, @tomvarsavsky, tweeted "One of the most interesting results I've seen in ML in the last 5 years. Evolving programs using a generic search space and generic mutations leads to the discovery of not only SGD and two layer NNs but also rand init, ReLU, Grad Norm. Can someone find a hidden inductive bias?".
Leading researcher Yoshua Bengio (Université de Montréal) published "Experience Grounds Language", which had 36 shares over the past 4 days. @pyoudeyer tweeted "On the topic of grounding language in action and social context, the work and roadmap made 10 years ago in development robotics (Italk EU project) is still fully relevant for #AI, #NPL and #DeepRL. Turn off the radio and go explore the world :)".
The paper shared the most on social media this week is by a team at Google: "Chip Placement with Deep Reinforcement Learning" by Azalia Mirhoseini et al (Apr 2020) with 523 shares. The authors present a learning - based approach to chip placement, one of the most complex and time - consuming stages of the chip design process. @ilmarihei (Ilmari Heikkinen) tweeted "Chip layouter makes 20% better layouts than team of experts, 100x faster. Interesting that human experts can get to within 20% of the layouter solution".
Over the past week, 192 new papers were published in "Computer Science - Computer Vision and Pattern Recognition".
The paper discussed most in the news over the past week was "Rapid AI Development Cycle for the Coronavirus (COVID-19) Pandemic: Initial Results for Automated Detection & Patient Monitoring using Deep Learning CT Image Analysis" by Ophir Gozes et al (Mar 2020), which was referenced 45 times, including in the article How a pandemic can present the opportunity for innovation in healthcare in Business LIVE South Africa. The paper author, Adam Bernheim (Icahn School of Medicine), was quoted saying "AI is of particular value in places where disease prevalence is high and availability of testing kits is limited". The paper got social media traction with 57 shares. A Twitter user, @primer_ai, said "The website identifies emerging topics, grouping research papers by common concepts. One topic is “Deep Learning & Machine Learning” with a popular study coming from that used AI-based image analysis tools for the detection of coronavirus".
Leading researcher Sergey Levine (University of California, Berkeley) came out with "Efficient Adaptation for End-to-End Vision-Based Robotic Manipulation" The researchers present a method and empirical evidence towards a robot learning framework that facilitates continuous adaption. @ericjang11 tweeted "If you can't tell if it is meta-learning or not, does it matter? 😋".
The paper shared the most on social media this week is "Unpaired Photo-to-manga Translation Based on The Methodology of Manga Drawing" by Hao Su et al (Apr 2020) with 2240 shares. The investigators propose MangaGAN, the first method based on Generative Adversarial Network (GAN) for unpaired photo - to - manga translation.
This week was active for "Computer Science - Computers and Society", with 32 new papers.
The paper discussed most in the news over the past week was by a team at Massachusetts Institute of Technology: "Apps Gone Rogue: Maintaining Personal Privacy in an Epidemic" by Ramesh Raskar et al (Mar 2020), which was referenced 15 times, including in the article MIT, Apple, Google Build Apps to Trace COVID-19 Contact in Government Technology US. The paper author, Ramesh Raskar (Massachusetts Institute of Technology), was quoted saying "We are dedicated to privacy-first solutions — user location and contact history should never leave a user’s phone without direct consent". The paper got social media traction with 96 shares. A user, @dhmackenzie, tweeted "Hi James, it's probably worth you taking a look first at this paper from the Safe Paths team, which explains some of the thinking. Safe Paths is a Privacy-First contact tracing app. Privacy has absolutely been a key consideration from day 1".
The paper shared the most on social media this week is "How Reliable are University Rankings?" by Ali Dasdan et al (Apr 2020) with 98 shares. @bianhaan (Dr. Bianca de Haan 🇬🇧🇪🇺🇳🇱#StayHomeSaveLives) tweeted "University rankings „we both formally and experimentally show in multiple ways that this ranking scheme is not reliable and cannot be trusted as authoritative because it is too sensitive to weight changes and can easily be gamed.“".
Over the past week, 15 new papers were published in "Computer Science - Human-Computer Interaction".
This week was very active for "Computer Science - Learning", with 287 new papers.
The paper discussed most in the news over the past week was "Rapid AI Development Cycle for the Coronavirus (COVID-19) Pandemic: Initial Results for Automated Detection & Patient Monitoring using Deep Learning CT Image Analysis" by Ophir Gozes et al (Mar 2020)
Leading researcher Yoshua Bengio (Université de Montréal) came out with "Experience Grounds Language", which had 36 shares over the past 4 days. @pyoudeyer tweeted "On the topic of grounding language in action and social context, the work and roadmap made 10 years ago in development robotics (Italk EU project) is still fully relevant for #AI, #NPL and #DeepRL. Turn off the radio and go explore the world :)".
The paper shared the most on social media this week is by a team at Google: "Chip Placement with Deep Reinforcement Learning" by Azalia Mirhoseini et al (Apr 2020)
Over the past week, ten new papers were published in "Computer Science - Multiagent Systems".
Over the past week, 30 new papers were published in "Computer Science - Neural and Evolutionary Computing".
The paper discussed most in the news over the past week was by a team at Google: "AutoML-Zero: Evolving Machine Learning Algorithms From Scratch" by Esteban Real et al (Mar 2020)
The paper shared the most on social media this week is "The Cost of Training NLP Models: A Concise Overview" by Or Sharir et al (Apr 2020) with 131 shares. @timFinin (Tim Finin) tweeted "This short paper from AI21 Labs ( gives an overview of the cost of training large neural network models for #NLP tasks (e.g., BERT, T5) which can run into millions of dollars and identifies some of the factors that impact those costs".
Over the past week, 34 new papers were published in "Computer Science - Robotics".
The paper discussed most in the news over the past week was "Improving Movement Predictions of Traffic Actors in Birds-Eye View Models using GANs and Differentiable Trajectory Rasterization" by Eason Wang et al (Apr 2020), which was referenced 4 times, including in the article Innovation Wrap: Shopify’s New Heights, Facebook Gaming, Dino DNA in ShareCafe. The paper got social media traction with 8 shares. On Twitter, @arXiv__ml posted "#machinelearning One of the most critical pieces of the self-driving puzzle is", while @eccdbb said "Improving Movement Predictions of Traffic Actors in Bird's-Eye View Models using GANs and Differentiable Trajectory Rasterization birds woke up what am i doing 5am".
Leading researcher Sergey Levine (University of California, Berkeley) published "Efficient Adaptation for End-to-End Vision-Based Robotic Manipulation" The researchers present a method and empirical evidence towards a robot learning framework that facilitates continuous adaption. @ericjang11 tweeted "If you can't tell if it is meta-learning or not, does it matter? 😋". This paper was also shared the most on social media with 139 tweets. @ericjang11 (Eric Jang 🇺🇸🇹🇼) tweeted "If you can't tell if it is meta-learning or not, does it matter? 😋".