Week Ending 07.14.19
RESEARCH WATCH: 07.14.19
Over the past week, 929 new papers were published in "Computer Science".
The paper discussed most in the news over the past week was by a team at Max Planck Institute for Informatics: "Text-based Editing of Talking-head Video" by Ohad Fried et al (Jun 2019), which was referenced 104 times, including in the article There's a terrifying trend on the internet that could be used to ruin your reputation, and no one knows how to stop it in Business Insider. The paper author, Maneesh Agrawala (Stanford University), was quoted saying "I think it's important to be careful and nuanced in how we talk about the potential for damage". The paper got social media traction with 28 shares. On Twitter, @michael_w_busch said "Checking the Fried et al. 2019 paper, there are artifacts in both video and audio that are obvious if you know what to look and listen for (the authors describe them as "an uncanny result"). Which does not erase the problems of weaponized disinformation".
Leading researcher Yoshua Bengio (Université de Montréal) came out with "Weakly-supervised Knowledge Graph Alignment with Adversarial Learning" The researchers propose to study aligning knowledge graphs in fully - unsupervised or weakly - supervised fashion, i.e., without or with only a few aligned triplets. @gastronomy tweeted "> This paper studies aligning knowledge graphs from different sources or languages. Most existing methods train supervised methods for the alig".
The paper shared the most on social media this week is by a team at Sorbonne Universités: "Large Memory Layers with Product Keys" by Guillaume Lample et al (Jul 2019) with 512 shares. @ylecun (Yann LeCun) tweeted "Awesome new paper from FAIR: 1. A new type of large-scale memory layer that uses product keys (FAISS-like indexing with product quantization) 2. Replace some layers in a BERT-like architecture by these Product Key Memory layers..... 3".
The most influential Twitter user discussing papers is Vinod Khosla who shared "Tackling Climate Change with Machine Learning" by David Rolnick et al (Jun 2019) and said: "Tackling climate change with AI? Unfortunately only "high risk" categories (& not all high risk ones) will make a material difference: fusion, materials/steel/cement, AV for public transit, alternative transit, alternative fuels, plant based food".
Over the past week, 65 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 Massachusetts Institute of Technology: "Tackling Climate Change with Machine Learning" by David Rolnick et al (Jun 2019), which was referenced 20 times, including in the article The AI Citizen and Climate Change in Medium.com. The paper author, David Rolnick(Massachusetts Institute of Technology), was quoted saying "Climate change does not present one problem, it presents multiple problems. AI is only one of the tools that can have an impact in the fight to mitigate the effects of climate change". The paper also got the most social media traction with 1439 shares. The researchers describe how machine learning can be a powerful tool in reducing greenhouse gas emissions and helping society adapt to a changing climate. On Twitter, @ikdeepl posted "Yikes “In 2006, at least two Scottish seafood firms flew hundreds of metric tons of shrimp from Scotland to China and Thailand for peeling, then back to Scotland for sale – because they could save on labor costs”".
Leading researcher Yoshua Bengio (Université de Montréal) came out with "Weakly-supervised Knowledge Graph Alignment with Adversarial Learning"
The paper shared the most on social media this week is by a team at Google: "Striving for Simplicity in Off-policy Deep Reinforcement Learning" by Rishabh Agarwal et al (Jul 2019) with 85 shares. @pcastr (Pablo Samuel Castro) tweeted "Neat paper showing you can train an #RL agent on Atari purely by sampling transitions from a DQN agent. This is very very off-policy, yet somehow seems to work! Great work led by , who recently joined our team in Montreal".
Over the past week, 175 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 Max Planck Institute for Informatics: "Text-based Editing of Talking-head Video" by Ohad Fried et al (Jun 2019)
Leading researcher Luc Van Gool (Computer Vision Laboratory) came out with "Learning a Curve Guardian for Motorcycles".
The paper shared the most on social media this week is by a team at Michigan State University: "R-Transformer: Recurrent Neural Network Enhanced Transformer" by Zhiwei Wang et al (Jul 2019) with 75 shares. @JamesONeil21 (JamesO'Neill) tweeted "Just like segmented recurrency in Trans-XL, recurrency still seems to play an important role".
The most influential Twitter user discussing papers is Vinod Khosla who shared "Tackling Climate Change with Machine Learning" by David Rolnick et al (Jun 2019)
Over the past week, 19 new papers were published in "Computer Science - Computers and Society".
The paper discussed most in the news over the past week was by a team at Massachusetts Institute of Technology: "Tackling Climate Change with Machine Learning" by David Rolnick et al (Jun 2019)
This week was active for "Computer Science - Human-Computer Interaction", with 31 new papers.
The paper discussed most in the news over the past week was by a team at Tsinghua University: "Identify and understand pay-it-forward reciprocity using millions of online red packets" by Yuan Yuan et al (Jun 2019), which was referenced 1 time, including in the article WeChat is running a natural experiment in human generosity in Technology Review. The paper author, Yuan, was quoted saying "Our natural experiment is enabled by the randomness in the mechanism that WeChat uses". The paper got social media traction with 11 shares.
This week was very active for "Computer Science - Learning", with 306 new papers.
The paper discussed most in the news over the past week was by a team at Max Planck Institute for Informatics: "Text-based Editing of Talking-head Video" by Ohad Fried et al (Jun 2019)
Leading researcher Yoshua Bengio (Université de Montréal) came out with "Weakly-supervised Knowledge Graph Alignment with Adversarial Learning"
The paper shared the most on social media this week is by a team at Sorbonne Universités: "Large Memory Layers with Product Keys" by Guillaume Lample et al (Jul 2019)
The most influential Twitter user discussing papers is Vinod Khosla who shared "Tackling Climate Change with Machine Learning" by David Rolnick et al (Jun 2019)
Over the past week, nine 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 Tartu: "Perspective Taking in Deep Reinforcement Learning Agents" by Aqeel Labashet al (Jul 2019), which was referenced 2 times, including in the article Last Week in AI in Hacker Noon. The paper got social media traction with 22 shares. The researchers present their progress toward building artificial agents with such abilities. A Twitter user, @dawn_alderson, commented "yes, a meta-learning strategy to close the gap between level 0 and level 2 perspective taking=RL has much promise but, I rather suspect hybridity needs to evolve in terms of the LSTM stack. hh", while @jaaanaru posted "Our first step toward creating AI agents with theory of mind: deep reinforcement learning agents can learn to take the perspective of another agent. This work was inspired by experiments done with chimpanzees, so we have monkeys and bananas on figures!".
Over the past week, 24 new papers were published in "Computer Science - Neural and Evolutionary Computing".
The paper shared the most on social media this week is "Sparse Networks from Scratch: Faster Training without Losing Performance" by Tim Dettmers et al (Jul 2019) with 126 shares. @MSripadarao (Manjunath Sripadarao) tweeted "Great paper, article excerpts, "...we can match or even exceed the performance of dense networks by using 20% of weights", and "...on cifar10 we can train both VGG-16-D and WRN-16-10 to dense performance levels with just 5% of weights", and the training is between 5x-12x faster!".
The most influential Twitter user discussing papers is Vinod Khosla who shared "Tackling Climate Change with Machine Learning" by David Rolnick et al (Jun 2019)
This week was active for "Computer Science - Robotics", with 61 new papers.
The paper discussed most in the news over the past week was by a team at University of Maryland: "Identifying Emotions from Walking using Affective and Deep Features" by Tanmay Randhavane et al (Jun 2019), which was referenced 6 times, including in the article 45 Numbers Highlighting The State Of AI Today in Forbes.com. The paper author, Aniket Bera (University of North Carolina), was quoted saying "Though we do not make any claims about the actual emotions a person is experiencing, our approach can provide an estimate of the perceived emotion of that walking style". The paper got social media traction with 184 shares. A user, @joemmac, tweeted "Gait is also sufficiently individualized that it can be used to ID targets with far less resolution and perfect angles than facial detection. And yes I used "targets" on purpose... welcome to the dystopia!".
Leading researcher Luc Van Gool (Computer Vision Laboratory) published "Learning a Curve Guardian for Motorcycles".