Week Ending 03.17.19
RESEARCH WATCH: 03.17.19
Over the past week, 206 new papers were published in "Computer Science".
The paper discussed most in the news over the past week was "Analyzing the Perceived Severity of Cybersecurity Threats Reported on Social Media" by Shi Zong et al (Feb 2019), which was referenced 12 times, including in the article Can Social Media Chatter Predict the Severity of Cybersecurity Threats? in EnterpriseCloudNews.com. The paper author, Alan Ritter (Ohio State University), was quoted saying "We think of it almost like Twitter trending topics". The paper got social media traction with 22 shares. The investigators investigate methods to analyze the severity of cybersecurity threats based on the language that is used to describe them online. A user, @sweetMals, tweeted "Classifying perceived severity of cyber security threats: ML approach. might be of your interest!", while @_kalavi_ commented "Interesting use of machine learning to determine severity of cyber threats".
Leading researcher Sergey Levine (University of California, Berkeley) published "Manipulation by Feel: Touch-Based Control with Deep Predictive Models" The investigators take a step towards addressing these issues by combining high - resolution tactile sensing with data - driven modeling using deep neural network dynamics models. @_derekwise_ tweeted "I'm fascinated by the problem of robotic manipulation of objects, partly because of its connection to the classical constraint problem of rolling without slipping---one of my favorite subjects. It's also interesting because humans do it so easily *by feel*".
The paper shared the most on social media this week is "To Tune or Not to Tune? Adapting Pretrained Representations to Diverse Tasks" by Matthew Peters et al (Mar 2019) with 239 shares. @penzant (Saku) tweeted "To Tune or Not to Tune? Adapting Pretrained Representations to Diverse Tasks neat emoji notation ❄ 🔥".
Over the past week, 68 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 3 times, including in the article What can psychopaths teach us about AI? in The Next Web. The paper got social media traction with 8 shares. The authors 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 Sergey Levine (University of California, Berkeley) came out with "Manipulation by Feel: Touch-Based Control with Deep Predictive Models" The authors take a step towards addressing these issues by combining high - resolution tactile sensing with data - driven modeling using deep neural network dynamics models. @_derekwise_ tweeted "I'm fascinated by the problem of robotic manipulation of objects, partly because of its connection to the classical constraint problem of rolling without slipping---one of my favorite subjects. It's also interesting because humans do it so easily *by feel*".
The paper shared the most on social media this week is "Can Robot Attract Passersby without Causing Discomfort by User-Centered Reinforcement Learning?" by Yasunori Ozaki et al (Mar 2019) with 84 shares.
Over the past week, 140 new papers were published in "Computer Science - Computer Vision and Pattern Recognition".
The paper discussed most in the news over the past week was "Predictive Inequity in Object Detection" by Benjamin Wilson et al (Feb 2019), which was referenced 63 times, including in the article Self-driving cars may hit people with darker skin more often in World Economic Forum. The paper author, Jamie Morgenstern (University of Pennsylvania), was quoted saying "The main takeaway from our work is that vision systems that share common structures to the ones we tested should be looked at more closely". The paper got social media traction with 162 shares. The investigators investigate whether state - of - the - art object detection systems have equitable predictive performance on pedestrians with different skin tones. A user, @defcon_5, tweeted "This is what systemic racism looks like.🚨🚨🚨 Black people may be at a greater risk of getting hit by self-driving cars because today's object-detection models exhibit higher precision on lighter skin tones #whitesupremacy".
Leading researcher Pieter Abbeel (University of California, Berkeley) published "Domain Randomization for Active Pose Estimation".
The paper shared the most on social media this week is by a team at University of California, Berkeley: "Manipulation by Feel: Touch-Based Control with Deep Predictive Models" by Stephen Tian et al (Mar 2019) with 71 shares. The researchers take a step towards addressing these issues by combining high - resolution tactile sensing with data - driven modeling using deep neural network dynamics models. @_derekwise_ (Derek Wise) tweeted "I'm fascinated by the problem of robotic manipulation of objects, partly because of its connection to the classical constraint problem of rolling without slipping---one of my favorite subjects. It's also interesting because humans do it so easily *by feel*".
Over the past week, 27 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 Cid:63)Cyprus University of Technology: "Disturbed YouTube for Kids: Characterizing and Detecting Disturbing Content on YouTube" by Kostantinos Papadamou et al (Jan 2019), which was referenced 29 times, including in the article YouTube says it bans preteens from its site. But it’s still delivering troubling content to young children. in Washington Post. The paper author, Jeremy Blackburn (University of Alabama at Birmingham), was quoted saying "I’m a parent. It’s definitely happening more often than I’m comfortable with". The paper got social media traction with 116 shares. A user, @a_papasavva, tweeted "Article by the about our work on detecting and characterizing disturbing YouTube videos for kids. We find that there is a 45% probability of kids reaching inappropriate content within 10 clicks when starting with a child-oriented video. 👌".
Over the past week, 18 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 3 times, including in the article What can psychopaths teach us about AI? in The Next Web. 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, 144 new papers were published in "Computer Science - Learning".
The paper discussed most in the news over the past week was "Predictive Inequity in Object Detection" by Benjamin Wilson et al (Feb 2019), which was referenced 63 times, including in the article Self-driving cars may hit people with darker skin more often in World Economic Forum. The paper author, Jamie Morgenstern (University of Pennsylvania), was quoted saying "The main takeaway from our work is that vision systems that share common structures to the ones we tested should be looked at more closely". The paper got social media traction with 162 shares. The authors investigate whether state - of - the - art object detection systems have equitable predictive performance on pedestrians with different skin tones. A Twitter user, @defcon_5, observed "This is what systemic racism looks like.🚨🚨🚨 Black people may be at a greater risk of getting hit by self-driving cars because today's object-detection models exhibit higher precision on lighter skin tones #whitesupremacy".
Leading researcher Sergey Levine (University of California, Berkeley) published "Manipulation by Feel: Touch-Based Control with Deep Predictive Models" The researchers take a step towards addressing these issues by combining high - resolution tactile sensing with data - driven modeling using deep neural network dynamics models. @_derekwise_ tweeted "I'm fascinated by the problem of robotic manipulation of objects, partly because of its connection to the classical constraint problem of rolling without slipping---one of my favorite subjects. It's also interesting because humans do it so easily *by feel*".
The paper shared the most on social media this week is "To Tune or Not to Tune? Adapting Pretrained Representations to Diverse Tasks" by Matthew Peters et al (Mar 2019) with 245 shares. @sleepinyourhat (Sam Bowman) tweeted "This is packed with useful experiments. I would probably have run half of them myself if this hadn't come out, and I probably wouldn't have been smart enough to think of the other half. Other reactions".
Over the past week, 11 new papers were published in "Computer Science - Multiagent Systems".
Over the past week, 19 new papers were published in "Computer Science - Neural and Evolutionary Computing".
The paper discussed most in the news over the past week was "SleepNet: Automated Sleep Disorder Detection via Dense Convolutional Neural Network" by Bahareh Pourbabaee et al (Mar 2019), which was referenced 2 times, including in the article AI classifies sleep disorders like sleep apnea, hypopnea, and arousal in Venturebeat. The paper got social media traction with 9 shares. On Twitter, @calogerozarbo observed "This is a really interesting domain where #DeepLearning has a lot to give. #machinelearning".
Leading researcher Kyunghyun Cho (New York University) published "Continual Learning via Neural Pruning".
This week was active for "Computer Science - Robotics", with 52 new papers.
The paper discussed most in the news over the past week was "Augmenting Visual SLAM with Wi-Fi Sensing For Indoor Applications" by Zakieh S. Hashemifar et al (Mar 2019), which was referenced 3 times, including in the article Wi-Fi data helps researchers improve accuracy of mapping algorithms in Venturebeat. The paper was shared 1 time in social media.
Leading researcher Sergey Levine (University of California, Berkeley) came out with "Manipulation by Feel: Touch-Based Control with Deep Predictive Models" The authors take a step towards addressing these issues by combining high - resolution tactile sensing with data - driven modeling using deep neural network dynamics models. @_derekwise_ tweeted "I'm fascinated by the problem of robotic manipulation of objects, partly because of its connection to the classical constraint problem of rolling without slipping---one of my favorite subjects. It's also interesting because humans do it so easily *by feel*".
The paper shared the most on social media this week is "Can Robot Attract Passersby without Causing Discomfort by User-Centered Reinforcement Learning?" by Yasunori Ozaki et al (Mar 2019) with 84 shares.