Week Ending 04.21.19
RESEARCH WATCH: 04.21.19
Over the past week, 270 new papers were published in "Computer Science".
The paper discussed most in the news over the past week was "Flash Boys 2.0: Frontrunning, Transaction Reordering, and Consensus Instability in Decentralized Exchanges"by Philip Daian et al (Apr 2019), which was referenced 25 times, including in the article The Ledger: China's Bitcoin Ban, Crypto Endowments, Arbitrage bots in Fortune. The paper author, Ariel Juels, was quoted saying "This should incentivize the community to consider new exchange designs". The paper got social media traction with 374 shares. On Twitter, @Miles_Kellerman commented "Interesting paper by et al. on front-running in dig. currencies: Can confirm from my field research that dig. curr. exchs VERY slow to set up surveillance. also find only 50% of top 10 have surveillance in place - shockingly low".
Leading researcher Yoshua Bengio (Université de Montréal) published "GradMask: Reduce Overfitting by Regularizing Saliency".
The paper shared the most on social media this week is by a team at Google: "NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection" by Golnaz Ghiasi et al (Apr 2019) with 424 shares. The investigators aim to learn a better architecture of feature pyramid network for object detection. @hardmaru (hardmaru) tweeted "Neural architecture search has been applied so far to (1) convnet-based image classifiers, (2) recurrent neural networks, (3) activation functions, (4) SGD optimizers, (5) data augmentation, (6) transformer, and now (7) object detection. What's next?".
Over the past week, 82 new papers were published in "Computer Science - Artificial Intelligence".
The paper discussed most in the news over the past week was "Improving Humanness of Virtual Agents and Users Cooperation through Emotions" by Moojan Ghafurian et al (Mar 2019), which was referenced 4 times, including in the article The strange way robots interact with psychopaths in TheLadders.com. The paper got social media traction with 8 shares. The investigators analyze the performance of an agent developed according to a well - accepted appraisal theory of human emotion with respect to how it modulates play in the context of a social dilemma.
Leading researcher Devi Parikh (Georgia Institute of Technology) came out with "Emergence of Compositional Language with Deep Generational Transmission".
The paper shared the most on social media this week is by a team at Stanford University: "4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks" by Christopher Choy et al (Apr 2019) with 129 shares. The investigators propose 4-dimensional convolutional neural networks for spatio - temporal perception that can directly process such 3D - videos using high - dimensional convolutions. @ArminMustafa (Armin Mustafa) tweeted "Great work on 3D video perception!".
This week was active for "Computer Science - Computer Vision and Pattern Recognition", with 268 new papers.
The paper discussed most in the news over the past week was "Vid2Game: Controllable Characters Extracted from Real-World Videos" by Oran Gafni et al (Apr 2019), which was referenced 17 times, including in the article Facebook AI turns real people into controllable game characters in Engadget UK. The paper got social media traction with 14 shares. On Twitter, @tonypeng_Synced said "In 's new study, researchers extract a controllable model from a video of a person performing a certain activity".
Leading researcher Yoshua Bengio (Université de Montréal) published "GradMask: Reduce Overfitting by Regularizing Saliency".
The paper shared the most on social media this week is "Fooling automated surveillance cameras: adversarial patches to attack person detection" by Simen Thys et al (Apr 2019) with 1639 shares. @pwang (Peter Wang) tweeted "This is going to be a major Tshirt trend over the next couple of years: Innocent-looking shirts with various patterns that are specifically designed to trick neural networks. Adversarial hats will also be a thing, for face detectors".
Over the past week, 23 new papers were published in "Computer Science - Computers and Society".
The paper discussed most in the news over the past week was "Discrimination through optimization: How Facebooks ad delivery can lead to skewed outcomes" by Muhammad Ali et al (Apr 2019), which was referenced 51 times, including in the article Discrimination's Digital Frontier in Atlantic.com - The Wire. The paper author, Alan Mislove(Computer science professor at Northeastern University), was quoted saying "All advertising is based on auctions all over the web, and I don’t know how you fix that without just saying we don’t have those kinds of ads". The paper also got the most social media traction with 1299 shares. The investigators demonstrate that such skewed delivery occurs on Facebook, due to market and financial optimization effects as well as the platforms own predictions about the relevance of ads to different groups of users. A Twitter user, @TimKarr, observed "April 4, 2019: Academic paper published analyzing FB's algorithms to find they're built using historically discriminatory data. The algorithms deliver results biased against people based on race & gender, & perpetuate discrimination in advertising".
The paper shared the most on social media this week is by a team at Stanford University: "From Theory to Systems: A Grounded Approach to Programming Language Education" by Will Crichton (Apr 2019) with 51 shares. @hn_frontpage (HN Front Page) tweeted "From Theory to Systems: A Grounded Approach to Programming Language Education L: C".
Over the past week, 21 new papers were published in "Computer Science - Human-Computer Interaction".
The paper discussed most in the news over the past week was "Improving Humanness of Virtual Agents and Users Cooperation through Emotions" by Moojan Ghafurian et al (Mar 2019), which was referenced 4 times, including in the article The strange way robots interact with psychopaths in TheLadders.com. The paper got social media traction with 8 shares. The investigators analyze the performance of an agent developed according to a well - accepted appraisal theory of human emotion with respect to how it modulates play in the context of a social dilemma.
Over the past week, 177 new papers were published in "Computer Science - Learning".
The paper discussed most in the news over the past week was "ClinicalBERT: Modeling Clinical Notes and Predicting Hospital Readmission" by Kexin Huang et al (Apr 2019), which was referenced 10 times, including in the article Researchers Develop AI To Predict Hospital Readmission Rates From Clinical Notes in Slashdot. The paper got social media traction with 31 shares. On Twitter, @NeuroWinter commented "Love seeing some real applications of BERT. It’s one thing seeing some amazing technology and it’s another seeing its uses", while @thejaan posted "Psyched to have helped a bit on this! ++ peep the funnest machine learning emoji 😂 Deep language models can predict patient readmission from clinical notes. Also don't miss concurrent work from".
Leading researcher Yoshua Bengio (Université de Montréal) came out with "GradMask: Reduce Overfitting by Regularizing Saliency".0
The paper shared the most on social media this week is by a team at Google: "NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection" by Golnaz Ghiasi et al (Apr 2019) with 424 shares. The researchers aim to learn a better architecture of feature pyramid network for object detection. @hardmaru (hardmaru) tweeted "Neural architecture search has been applied so far to (1) convnet-based image classifiers, (2) recurrent neural networks, (3) activation functions, (4) SGD optimizers, (5) data augmentation, (6) transformer, and now (7) object detection. What's next?".
Over the past week, seven new papers were published in "Computer Science - Multiagent Systems".
The paper discussed most in the news over the past week was "From Motions to Emotions: Can the Fundamental Emotions be Expressed in a Robot Swarm?" by María Santoset al (Mar 2019), which was referenced 1 time, including in the article From motion to emotion: The potential of robot swarms in artistic performances in PhysOrg.com. The paper author, Maria Santos, was quoted saying "Given that the robots in the swarm are very simple and do not have a face or a body, it is remarkable that these simple behaviors are directly identified by the spectators as conveying the intended emotions". The paper was shared 4 times in social media. The investigators explore the expressive capabilities of a swarm of miniature mobile robots within the context of inter - robot interactions and their mapping to the so - called fundamental emotions.
Over the past week, 30 new papers were published in "Computer Science - Neural and Evolutionary Computing".
This week was active for "Computer Science - Robotics", with 59 new papers.
The paper discussed most in the news over the past week was "Real-Time Dense Stereo Embedded in A UAV for Road Inspection" by Rui Fan et al (Apr 2019), which was referenced 2 times, including in the article Team developing AI-enabled drones for pothole, crack detection in GlobalSpec. The paper got social media traction with 6 shares. The authors present a robust stereo vision system embedded in an unmanned aerial vehicle (UAV).
Leading researcher Sergey Levine (University of California, Berkeley) came out with "End-to-End Robotic Reinforcement Learning without Reward Engineering" The authors propose an approach for removing the need for manual engineering of reward specifications by enabling a robot to learn from a modest number of examples of successful outcomes, followed by actively solicited queries, where the robot shows the user a state and asks for a label to determine whether that state represents successful completion of the task. @snowy_robolamp tweeted "Impressive speed of learning! 🤖". This paper was also shared the most on social media with 100 tweets. @snowy_robolamp (ЮляN8FAD85042👩💻🐧❄️) tweeted "Impressive speed of learning! 🤖".