Week Ending 2.27.2022
RESEARCH WATCH: 2.27.2022
SPONSORED BY
Over the past week, 1,013 new papers were published in "Computer Science".
The paper discussed most in the news over the past week was "Obstacle avoidance for blind people using a 3D camera and a haptic feedback sleeve" by Manuel Zahn et al (Jan 2022), which was referenced 31 times, including in the article This Device Lets Blind People See With Vibrations In Their Arms in Tech Register. The paper got social media traction with 8 shares. A Twitter user, @stripp_lab, commented "Quite some media coverage on this 'infrared vision' pre-print... Data not convincing though".
Leading researcher Sergey Levine (University of California, Berkeley) published "ViKiNG: Vision-Based Kilometer-Scale Navigation with Geographic Hints" The researchers propose a learning - based approach that integrates learning and planning, and can utilize side information such as schematic roadmaps, satellite maps and GPS coordinates as a planning heuristic, without relying on them being accurate.
The paper shared the most on social media this week is by a team at Skolkovo Institute of Science and Technology: "Survey on Large Scale Neural Network Training" by Julia Gusak et al (Feb 2022) with 122 shares. @FinSentim (FinSentim) tweeted "This survey provides a systematic review of the approaches that enable more efficient DNNs training. Julia Gusak et al. analyze techniques that save memory and make good use of computation. They summarize the main categories of strategies and compare strategies within categories".
The most influential Twitter user discussing papers is AK who shared "Designing Effective Sparse Expert Models" by Barret Zoph et al (Feb 2022) and said: "Designing Effective Sparse Expert Models abs: A 269B sparse model (the Stable Transferable Mixture-of-Experts or ST-MoE-32B) which achieves sota performance across a diverse set of natural language benchmarks".
This week was very active for "Computer Science - Artificial Intelligence", with 177 new papers.
The paper discussed most in the news over the past week was by a team at Microsoft: "Singularity: Planet-Scale, Preemptive and Elastic Scheduling of AI Workloads" by Dharma Shukla (Microsoft) et al (Feb 2022), which was referenced 20 times, including in the article Microsoft goes public with details on its 'Singularity' AI infrastructure service in ZDNet. The paper got social media traction with 139 shares. On Twitter, @CKsTechNews posted "#Microsoft details 'planet-scale' AI infrastructure packing 100k-plus #GPUs Microsoft with the big mouth, less talk, more doing friends. Press Paper", while @PaperTldr said "🗜87% Scheduling high utilization across deep learning and inference workloads is a crucial lever for cloud providers to train and deliver highly-efficient distributed service to their clients".
Leading researcher Sergey Levine (University of California, Berkeley) came out with "ViKiNG: Vision-Based Kilometer-Scale Navigation with Geographic Hints" The researchers propose a learning - based approach that integrates learning and planning, and can utilize side information such as schematic roadmaps, satellite maps and GPS coordinates as a planning heuristic, without relying on them being accurate.
The paper shared the most on social media this week is by a team at Skolkovo Institute of Science and Technology: "Survey on Large Scale Neural Network Training" by Julia Gusak et al (Feb 2022)
Over the past week, 201 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 Deakin University: "Towards Effective and Robust Neural Trojan Defenses via Input Filtering" by Kien Do et al (Feb 2022), which was referenced 3 times, including in the article Techniques to fool AI with hidden triggers are outpacing defenses – study in TheRegister.com. The paper got social media traction with 6 shares. On Twitter, @summarizedml observed "New defenses that can mitigate several Trojan attacks and improve on the state-of-the-art defenses. 📄".
The paper shared the most on social media this week is by a team at Google: "Self-Distilled StyleGAN: Towards Generation from Internet Photos" by Ron Mokady et al (Feb 2022) with 78 shares. The researchers show how StyleGAN can be adapted to work on raw uncurated images collected from the Internet. @xsteenbrugge (Xander Steenbrugge) tweeted "Everyone using diffusion models: "GANs are dead cause they have bad coverage of the data manifold" GAN researchers: "Hold on, hold on... This ain't over!"".
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 "AirGuard -- Protecting Android Users From Stalking Attacks By Apple Find My Devices" by Alexander Heinrich et al (Feb 2022), which was referenced 2 times, including in the article AirGuard Android app offers better detection of AirTag stalking in Bleeping Computer. The paper got social media traction with 22 shares. On Twitter, @maggied said "Have not dug into it yet but just saw this paper about AirGuard that they mention in the positive security post Also Re tracker detect is the AirTag paired or unpaired w Apple account? It worked for me with a paired one so long as out of range of owner".
This week was active for "Computer Science - Human-Computer Interaction", with 30 new papers.
The paper discussed most in the news over the past week was "Obstacle avoidance for blind people using a 3D camera and a haptic feedback sleeve" by Manuel Zahn et al (Jan 2022)
This week was very active for "Computer Science - Learning", with 400 new papers.
The paper discussed most in the news over the past week was by a team at UC Berkeley: "ViKiNG: Vision-Based Kilometer-Scale Navigation with Geographic Hints" by Dhruv Shah et al (Feb 2022), which was referenced 3 times, including in the article Google-Berkeley robot navigation could chart a new-course for self-driving systems in ZDNet. The paper author, Sergey Levine (University of California, Berkeley), was quoted saying "But autonomous driving or other tasks with higher stakes (or even real sidewalk delivery that has to deal with dense traffic) has to have additional mechanisms to deal with safety and constraints, which the current approach doesn't directly handle just yet". The paper got social media traction with 20 shares. The authors propose a learning - based approach that integrates learning and planning, and can utilize side information such as schematic roadmaps, satellite maps and GPS coordinates as a planning heuristic, without relying on them being accurate.
The paper shared the most on social media this week is by a team at Skolkovo Institute of Science and Technology: "Survey on Large Scale Neural Network Training" by Julia Gusak et al (Feb 2022)
The most influential Twitter user discussing papers is AK who shared "Designing Effective Sparse Expert Models" by Barret Zoph et al (Feb 2022)
This week was active for "Computer Science - Multiagent Systems", with 22 new papers.
Over the past week, 19 new papers were published in "Computer Science - Neural and Evolutionary Computing".
Leading researcher Quoc V. Le (Google) came out with "Transformer Quality in Linear Time" @HochreiterSepp tweeted "ArXiv Fast linear transformer attention using gated linear (attention) units and mixed chunk attention with input chunked. Training speedups of 4.9 on Wiki-40B and 12.1 on PG-19 for auto-regressive and 4.8 on C4 for masked language modeling".
This week was very active for "Computer Science - Robotics", with 87 new papers.
The paper discussed most in the news over the past week was by a team at UC Berkeley: "ViKiNG: Vision-Based Kilometer-Scale Navigation with Geographic Hints" by Dhruv Shah et al (Feb 2022)