Week Ending 05.19.19
RESEARCH WATCH: 05.19.19
Over the past week, 929 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 56 times, including in the article Bancor had a front-running problem, so it hired one of the manipulators to fix it in Yahoo! Finance. The paper author, Philip Daian, was quoted saying "The miners have a tremendous amount of power". The paper got social media traction with 396 shares. A Twitter user, @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) came out with "GMNN: Graph Markov Neural Networks" The investigators study semi - supervised object classification in relational data, which is a fundamental problem in relational data modeling.
The paper shared the most on social media this week is by a team at DeepMind: "Deep Compressed Sensing" by Yan Wu et al (May 2019) with 176 shares. @__MLT__ (Machine Learning Tokyo) tweeted "Paper to read on the weekend? 🤩".
This week was active for "Computer Science - Artificial Intelligence", with 98 new papers.
The paper discussed most in the news over the past week was "Magic: The Gathering is Turing Complete" by Alex Churchill et al (Mar 2019), which was referenced 16 times, including in the article Magic: The Gathering is World's 'Most Complex' Game, Study Finds in MSNBC Newsweek. The paper author, Alex Churchill, was quoted saying "I was on a message board when someone asked whether Magic: The Gathering was Turing complete". The paper also got the most social media traction with 493 shares. The authors show that optimal play in real - world Magic Magic is at least as hard as the Halting Problem, solving a problem that has been open for a decade. A Twitter user, @TomRivlin, commented "It's a good day today for scientific papers about games of various sorts", while @jackmcgraf commented "This is the first paper that's made me think grad school could be fun. This is so good. Proving Magic: The Gathering is Turing Complete".
Leading researcher Pieter Abbeel (University of California, Berkeley) came out with "Bit-Swap: Recursive Bits-Back Coding for Lossless Compression with Hierarchical Latent Variables" @gastronomy tweeted "> The bits-back argument suggests that latent variable models can be turned into lossless compression schemes. Trans".
The paper shared the most on social media this week is "Learnable Triangulation of Human Pose" by Karim Iskakov et al (May 2019) with 101 shares. @DmitryUlyanovML (Dmitry Ulyanov) tweeted "Check out a new paper from my colleagues at Samsung! They smashed previous SOTA in 3D human pose estimation using a novel multi-view volumetric aggregation method. ▶️ 🌐 📝".
Over the past week, 187 new papers were published in "Computer Science - Computer Vision and Pattern Recognition".
The paper discussed most in the news over the past week was by a team at Massachusetts Institute of Technology: "Adversarial Examples Are Not Bugs, They Are Features"by Andrew Ilyas et al (May 2019), which was referenced 18 times, including in the article admin wrote a new post, Scientists help artificial intelligence outsmart hackers | Science in DigitalMunition. The paper author, Shibani Santurkar (Massachusetts Institute of Technology), was quoted saying "If we know that our models are relying on these microscopic patterns that we don’t see, then we can’t pretend that they are interpretable in a human fashion". The paper got social media traction with 530 shares. A Twitter user, @SashaVNovikov, observed "Cool! Like furry ear is a feature which can be used to detect cats, adversarial perturbations are features of natural images which can be used to correctly classify both train and test data, except humans don't see it. So adversarial perturbations are human's bugs, not model's".
Leading researcher Pieter Abbeel (University of California, Berkeley) came out with "Population Based Augmentation: Efficient Learning of Augmentation Policy Schedules" The researchers introduce a new data augmentation algorithm, Population Based Augmentation (PBA), which generates nonstationary augmentation policy schedules instead of a fixed augmentation policy.
The paper shared the most on social media this week is "Learnable Triangulation of Human Pose" by Karim Iskakov et al (May 2019) with 101 shares. @DmitryUlyanovML (Dmitry Ulyanov) tweeted "Check out a new paper from my colleagues at Samsung! They smashed previous SOTA in 3D human pose estimation using a novel multi-view volumetric aggregation method. ▶️ 🌐 📝".
Over the past week, 28 new papers were published in "Computer Science - Computers and Society".
The paper discussed most in the news over the past week was "Large-scale and high-resolution analysis of food purchases and health outcomes" by Luca Maria Aiello et al (Apr 2019), which was referenced 3 times, including in the article Grocery bills can predict diabetes rates by neighborhood in Technology Review. The paper author, Luca Maria Aiello, was quoted saying "The residents of Hackney, which is a deprived yet highly-educated neighbourhood in East London, enjoy healthier eating habits and do not suffer from diabetes as much as Newham’s residents". The paper got social media traction with 6 shares.
This week was active for "Computer Science - Human-Computer Interaction", with 32 new papers.
The paper discussed most in the news over the past week was by a team at University of Pennsylvania: "What Twitter Profile and Posted Images Reveal About Depression and Anxiety" by Sharath Chandra Guntuku et al (Apr 2019), which was referenced 22 times, including in the article Twitter image colors and content could help identify users with depression, anxiety in EurekAlert!. The paper author, Sharath Guntuku, was quoted saying "It is challenging to transform pixels that form the images to interpretable features, but with the advances in computer vision algorithms, we are now attempting to uncover another dimension of the condition as it manifests online." The paper got social media traction with 9 shares. The researchers examine which attributes of profile and posted images are associated with depression and anxiety of Twitter users.
Leading researcher Jianfeng Gao (Microsoft) came out with "Challenges in Building Intelligent Open-domain Dialog Systems".
This week was very active for "Computer Science - Learning", with 343 new papers.
The paper discussed most in the news over the past week was by a team at Google: "Direct speech-to-speech translation with a sequence-to-sequence model" by Ye Jia et al (Apr 2019), which was referenced 31 times, including in the article Google’s Translatotron converts one spoken language to another, no text involved in TechCrunch. The paper got social media traction with 55 shares. A Twitter user, @bryanoloughlin, said "Translatotron : An End-to-End Speech-to-Speech Translation Model , GoogleAI May 15, 2019 1st-of-its-kind,work-in-prog,mimics users own voice, speed Direct speech-to-speech translation with a seq 2-seq model".
Leading researcher Yoshua Bengio (Université de Montréal) published "GMNN: Graph Markov Neural Networks" The researchers study semi - supervised object classification in relational data, which is a fundamental problem in relational data modeling.
The paper shared the most on social media this week is by a team at DeepMind: "Deep Compressed Sensing" by Yan Wu et al (May 2019) with 176 shares. @__MLT__ (Machine Learning Tokyo) tweeted "Paper to read on the weekend? 🤩".
Over the past week, 12 new papers were published in "Computer Science - Multiagent Systems".
The paper discussed most in the news over the past week was by a team at University of California, Berkeley: "Behavior Planning of Autonomous Cars with Social Perception"by Liting Sun et al (May 2019), which was referenced 1 time, including in the article A social perception scheme for behavior planning of autonomous cars in PhysOrg.com. The paper got social media traction with 6 shares. The authors propose a social perception scheme which treats all road participants as distributed sensors in a sensor network.
Over the past week, 20 new papers were published in "Computer Science - Neural and Evolutionary Computing".
This week was active for "Computer Science - Robotics", with 49 new papers.
The paper discussed most in the news over the past week was "Bee++: A 95-mg Four-Winged Insect-Scale Flying Robot Driven by Twinned Unimorph Actuators" by Xiufeng Yang et al (May 2019), which was referenced 7 times, including in the article Scientists create a four-winged robot insect that flies with grace in Engadget UK. The paper was shared 3 times in social media. On Twitter, @bwaber posted "Fascinating: we can make robotic dragonflies now".
Leading researcher Trevor Darrell (UC Berkeley) published "Monocular Plan View Networks for Autonomous Driving".