Week Ending 3.7.2021
RESEARCH WATCH: 3.7.2021
This week was active for "Computer Science", with 1,092 new papers.
The paper discussed most in the news over the past week was by a team at University of Oxford: "QNLP in Practice: Running Compositional Models of Meaning on a Quantum Computer" by Robin Lorenz et al (Feb 2021), which was referenced 61 times, including in the article Cambridge Quantum Announce Largest Ever Natural Language Processing Implementation on a Quantum Computer in Canada NewsWire. The paper got social media traction with 37 shares. The researchers present results on the first NLP experiments conducted on Noisy Intermediate - Scale Quantum (NISQ) computers for datasets of size >= 100 sentences. A Twitter user, @Moor_Quantum, observed "Very nice work by team that advances QNLP and provides significant POC that it can be done on quantum. Covers tech problems and training issues of running a NPL model with >100 sentences on NISQ QC".
Leading researcher Yoshua Bengio (Université de Montréal) published "Neural Production Systems" @bengoertzel tweeted "nice progress from Yoshua Bengio etc. on extracting symbolic rules from data using neural nets ... limited in scope/power but promising".
The paper shared the most on social media this week is "Neural 3D Video Synthesis" by Tianye Li et al (Mar 2021) with 275 shares. @rmbrualla (Ricardo Martin-Brualla) tweeted "Really cool work from my neighbor Mira Slavcheva!! 😃 The cooking dataset makes me *so hungry*! Seems like it is a pandemic-inspired dataset, like the ones we got for nerfies. Congrats y'all! #futureofcooking".
The most influential Twitter user discussing papers is Sebastian Raschka who shared "Generative Adversarial Transformers" by Drew A. Hudson et al (Mar 2021) and said: "GANsformer: 2 weeks later, the second methods on GANs with Transformers just went live: Wow, their generated images look better than anything I was ever able to generate with existing GANs".
This week was very active for "Computer Science - Artificial Intelligence", with 150 new papers.
The paper discussed most in the news over the past week was by a team at University of Oxford: "QNLP in Practice: Running Compositional Models of Meaning on a Quantum Computer" by Robin Lorenz et al (Feb 2021)
Leading researcher Yoshua Bengio (Université de Montréal) came out with "Neural Production Systems" @bengoertzel tweeted "nice progress from Yoshua Bengio etc. on extracting symbolic rules from data using neural nets ... limited in scope/power but promising".
The paper shared the most on social media this week is by a team at Stanford University: "Generative Adversarial Transformers" by Drew A. Hudson et al (Mar 2021) with 226 shares. @artsiom_s (Artsiom Sanakoyeu) tweeted "Generative Adversarial Transformers 📝 🛠️ The GANsformer leverages a bipartite structure to allow long-range interactions, while evading the quadratic complexity standard transformers suffer from. Presented 2 novel attention types".
This week was active for "Computer Science - Computer Vision and Pattern Recognition", with 257 new papers.
The paper discussed most in the news over the past week was by a team at DeepMind: "High-Performance Large-Scale Image Recognition Without Normalization" by Andrew Brock et al (Feb 2021), which was referenced 9 times, including in the article Best of arXiv — Readings for March 2021 in Towards Data Science. The paper got social media traction with 1085 shares. The investigators develop an adaptive gradient clipping technique which overcomes these instabilities, and design a significantly improved class of Normalizer - Free ResNets. A user, @sohamde_, tweeted "Releasing NFNets: SOTA on ImageNet. Without normalization layers! Code: This is the third paper in a series that began by studying the benefits of BatchNorm and ended by designing highly performant networks w/o it. A thread: 1/8".
The paper shared the most on social media this week is "Neural 3D Video Synthesis" by Tianye Li et al (Mar 2021)
The most influential Twitter user discussing papers is Sebastian Raschka who shared "Generative Adversarial Transformers" by Drew A. Hudson et al (Mar 2021)
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 "Auditing E-Commerce Platforms for Algorithmically Curated Vaccine Misinformation" by Prerna Juneja et al (Jan 2021), which was referenced 7 times, including in the article Amazon directs customers to vaccine misinformation, study finds in EuroNews. The paper author, Prerna Juneja, was quoted saying "We found out that once users start engaging with misinformation on the platform, they are presented with more misinformation at various points in their Amazon navigation route". The paper got social media traction with 27 shares. A Twitter user, @JMarkOckerbloom, observed "Link to preprint discussed in the story: Among the findings are that misinforming books often get ranked higher than books that debunk them, and that you get notably more misinformation recommendations once you click on one of them (even if you don't buy)".
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 "Auditing E-Commerce Platforms for Algorithmically Curated Vaccine Misinformation" by Prerna Juneja et al (Jan 2021)
This week was very active for "Computer Science - Learning", with 373 new papers.
The paper discussed most in the news over the past week was by a team at University of Oxford: "QNLP in Practice: Running Compositional Models of Meaning on a Quantum Computer" by Robin Lorenz et al (Feb 2021)
Leading researcher Yoshua Bengio (Université de Montréal) published "Neural Production Systems" @bengoertzel tweeted "nice progress from Yoshua Bengio etc.
The paper shared the most on social media this week is by a team at Stanford University: "Generative Adversarial Transformers" by Drew A. Hudson et al (Mar 2021)
Over the past week, 13 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 Google: "Adversarial Environment Generation for Learning to Navigate the Web" by Izzeddin Gur et al (Mar 2021), which was referenced 1 time, including in the article PAIRED: A New Multi-agent Approach for Adversarial Environment Generation in Google AI Blog. The paper was shared 1 time in social media. A user, @gastronomy, tweeted "> Learning to autonomously navigate the web is a difficult sequential decision making task. The state and action spaces are large and combinatori".
Over the past week, 23 new papers were published in "Computer Science - Neural and Evolutionary Computing".
The paper discussed most in the news over the past week was by a team at Google: "Evolving Reinforcement Learning Algorithms" by John D. Co-Reyes et al (Jan 2021), which was referenced 3 times, including in the article Implementing DQNClipped and DQNReg with Stable Baselines in Medium.com. The paper got social media traction with 81 shares. A Twitter user, @AlifePapers, said "EVOLVING REINFORCEMENT LEARNING ALGORITHMS "We propose a method for meta-learning reinforcement learning algorithms by searching over the space of computational graphs which compute the loss function for a value-based model-free RL agent to optimize"".
This week was very active for "Computer Science - Robotics", with 103 new papers.
The paper discussed most in the news over the past week was by a team at Google: "How to Train Your Robot with Deep Reinforcement Learning; Lessons Weve Learned" by Julian Ibarz et al (Feb 2021), which was referenced 2 times, including in the article Way beyond AlphaZero: Berkeley and Google work shows robotics may be the deepest machine learning of all in ZDNet. The paper author, Sergey Levine (University of California, Berkeley), was quoted saying "and it's one of the places where the standard RL [reinforcement learning] problem statement, which assumes the reward is simply 'provided' somehow to the agent (e.g., with a piece of code), deviates from the requirements in the real world." The paper got social media traction with 310 shares. On Twitter, @svlevine observed "What did we learn from 5 years of robotic deep RL? My colleagues at Google and I tried to distill our experience into a review-style journal paper, covering some of the practical aspects of real-world robotic deep RL: 🧵->".