News & trends in the Last 48 hours
News & trends in the Last 48 hours
In the Last 48 hours there's been 10 influencers like ylecun and hardmaru, discussing topics such as #AI, #BreastCancerand #CoNLL2019.
Trends
Top hashtags
Top influencers
ylecun 
hardmaru 
stanfordnlp 
weballergy 
jeremyphoward 
peteskomoroch 
NandoDF 
adriancolyer 
lawrennd Top sources
News
An Exponential Learning Rate Schedule for Deep Learning
RT @prfsanjeevarora: Conventional wisdom: slowly decay learning rate (lr) when training deep nets. Empirically, some exotic lr schedules also work, eg cosine. New work with Zhiyuan Li: exponentially increasing lr works too! Experiments + surprising math explanation. See https://t.co/wTCUZjQ8jV
OpenIntriguing empirical evidence exists that deep learning can work well with exoticschedules for varying the learning rate. This paper suggests that the phenomenon may be due to Batch Normalization or BN, which is ubiquitous and provides benefits in optimization and generalization across ...
RT @prfsanjeevarora: Conventional wisdom: slowly decay learning rate (lr) when training deep nets. Empirically, some exotic lr schedules also work, eg cosine. New work with Zhiyuan Li: exponentially increasing lr works too! Experiments + surprising math explanation. See https://t.co/wTCUZjQ8jV
OpenAn Exponential Learning Rate Schedule for Deep Learning
Intriguing empirical evidence exists that deep learning can work well with exoticschedules for varying the learning rate. This paper suggests that the phenomenon may be due to Batch ...
RT @roadrunning01: Visual Hide and Seek
pdf: https://t.co/dbxvdPwZVl
abs: https://t.co/KSsU8hQQKN
project page: https://t.co/avKeboftEA https://t.co/gasqhV92dZ
OpenClick here to read the article
Hi er SeekerVisual Observation Visual Observation Hider Seeker (c) Hider’s Visual Observation(a) Seeker’s Visual Observation (b) Simulation Environment Figure 1: Visual Hide-and-Seek: We ...
RT @cai2r: Our #AI improves #BreastCancer detection.
Technical ℹ️
[Paper 🆕] https://t.co/Yei5PiYfh8
[Code] https://t.co/yTUapbgQog
[Data specs] https://t.co/m058wcKfac
Lay-friendly ℹ️
[Press release🆕] https://t.co/22cxiKzvmx
[Explainer] https://t.co/lIUVoavOw3
@NanWu__ @kjgeras & al. https://t.co/799DX6kKHu
OpenCombination of Artificial Intelligence & Radiologists More Accurately Identified Breast Cancer
NYU School of Medicine researchers found their AI tool recognized patterns in screening mammogram images radiologists could not, and vice versa. Learn more.
RT @cai2r: Our #AI improves #BreastCancer detection.
Technical ℹ️
[Paper 🆕] https://t.co/Yei5PiYfh8
[Code] https://t.co/yTUapbgQog
[Data specs] https://t.co/m058wcKfac
Lay-friendly ℹ️
[Press release🆕] https://t.co/22cxiKzvmx
[Explainer] https://t.co/lIUVoavOw3
@NanWu__ @kjgeras & al. https://t.co/799DX6kKHu
OpenDeep Neural Networks Improve Radiologists’ Performance in Breast Cancer Screening
Using a new massive mammography dataset, we trained a model that is as accurate as radiologists in classifying breast cancer screening…
RT @bhutanisanyam1: I’m really honoured to share my interview with @chipro from @nvidia all about Machine Learning interviews, DL Reseach, MOOCs and ofcourse Chip’s upcoming book:
Audio: https://t.co/vRWXU6HVVz
Video: https://t.co/CDVeo16P5c
OpenChip Huyen Interview: Machine Learning Interviews | MOOCS and Deep Learning at NVIDIA by Chai Time Data Science • A podcast on Anchor
Audio (Podcast Version) available here: https://anchor.fm/chaitimedatascience Subscribe here to the newsletter: https://tinyletter.com/sanyambhutani In this episode, Sanyam Bhutani ...
RT @maalvarezl: I'm advertising a PhD position on Automatic Learning of Latent Force Models at @shefcompsci. Learning physics from data seems to be a new trend in ML. We'll do it in the context of Gaussian processes. https://t.co/pKkNdjA6W2
OpenAutomatic Learning of Latent Force Models
PhD Project - Automatic Learning of Latent Force Models at University of Sheffield, listed on FindAPhD.com
Improving Gradient Estimation in Evolutionary Strategies With Past Descent Directions
https://t.co/jxIIft8Bc3 https://t.co/c9yJUjWpFb
OpenImproving Gradient Estimation in Evolutionary Strategies With Past Descent Directions
Evolutionary Strategies (ES) are known to be an effective black-box optimization technique for deep neural networks when the true gradients cannot be computed, such as in Reinforcement ...
RT @arokem: @allafarce @peteskomoroch @jeremyphoward There's @akeshavan_ 's (open source!/free!) Swipes for Science: https://t.co/SLZu1uoIjr. We demonstrated how it works in a paper you can read here: https://t.co/BFJjc7qmV9
OpenCombining Citizen Science and Deep Learning to Amplify Expertise in Neuroimaging
Big Data promises to advance science through data-driven discovery. However, many standard lab protocols rely on manual examination, which is not feasible for large-scale datasets. ...
RT @MicrochipMakes: Run a machine learning model on an @adafruit Trinket M0 to recognize gestures: https://t.co/rGLqpUTUdb https://t.co/N6ohbyoVn0
OpenMACHINE LEARNING MONDAY – Machine Learning on an Adafruit Trinket M0 to recognize gestures @adafruit @Microsoft #ProtoNN @Paulfo #TheThingsConference @TTN_UK
GesturePod – A workshop on using EdgeML ProtoNN to create a gesture recognising device on constrained hardware – GesturePod – A workshop on using EdgeML ProtoNN to create a gestur…
@allafarce @jeremyphoward For using your own internal “workers” to label images:
LocalTurk https://t.co/R7qawAczwB, @labelbox, @FigureEightInc, @amazonmturk with private qualification, https://t.co/GtlOuU1e0J
Openlocalturk
Mechanical Turk on your own machine. Contribute to danvk/localturk development by creating an account on GitHub.
Browse topics
Get updates live! Follow us on
RT @_KarenHao: IT'S HERE!!! The biggest story I’ve ever worked on. @techreview's very first interactive ever, which walks through a concrete example of AI bias, and why it’s so much more complicated than initially meets the eye. https://t.co/4eqXB3kpFF 1/
OpenCan you make AI fairer than a judge? Play our courtroom algorithm game
The US criminal legal system uses predictive algorithms to try to make the judicial process less biased. But there’s a deeper problem.
"HackPPL: a universal probabilistic programming language" Ai et al., MAPL'19 https://t.co/ZYMUK2a9nR #themorningpaper
Facebook have extended Hack with probabilistic programming support to better support reasoning over uncertainty in their application logic. https://t.co/QknqAjeoBf
OpenHackPPL: a universal probabilistic programming language
HackPPL: a universal probabilistic programming language Ai et al., MAPL’19 The Hack programming language, as the authors proudly tell us, is “a dominant web development language across …
RT @razielag: Pretty cool use of #TensorFlow
@petewarden this seems like another great fit for TF Micro.
Using Old Cellphones to Listen for Illegal Loggers https://t.co/v85Wsf33FI
OpenUsing Old Cellphones to Listen for Illegal Loggers
Indonesian villagers are trying out a treetop surveillance system that uses recycled phones and artificial intelligence software to detect chain saws.