Project Jupyter is a huge hit in data science, but it has not yet found widespread adoption in robotics. Today, we are releasing the first version of jupyter-ros, a collection of Jupyter interactive widgets inspired by Qt and RViz, to bring their features to the Jupyter ecosystem. This may be the right time for Jupyter-based developer tools, as cloud robotics is taking off."
Blog by Wolf Vollprecht:

https://blog.jupyter.org/ros-jupyter-b7e82b5e1202


#Robotics #Python #RobotOperatingSystem #Visualization #CloudRobotics

The power of deeper networks for expressing natural functions
David Rolnick & Max Tegmark: https://arxiv.org/abs/1705.05502
#DeepLearning #MachineLearning #NeuralComputing

Music Transformer


arxiv.org/abs/1809.04281

#DeepLearning #Transformer #MachineLearning #SpeechProcessing #Music

Reinforcement Learning with Attention that Works: A Self-Supervised Approach"
Manchin et al.: https://arxiv.org/abs/1904.03367

"That’s where an algorithm can help: once trained, it could reliably catch congenital heart disease in perpetuity. Catching heart defects early can lead to better outcomes for patients after birth. And if certain types of lesions are spotted in a fetal ultrasound, doctors can recommend in-utero therapies that significantly improve the heart’s condition by birth."

https://blogs.nvidia.com/blog/2019/03/21/ucsf-heart-defects-ai/?

The quantum physics community makes the ML community look like a bunch of beginners.
While we're arguing about the importance of reproducibility they *experimentally prove that there is no such thing as observer-independent objective truth*" - Ferenc Huszar
A quantum experiment suggests there’s no such thing as objective reality:
https://www.technologyreview.com/s/613092/a-quantum-experiment-suggests-theres-no-such-thing-as-objective-reality/
#MachineLearning #Physics #QuantumMechanics

Learning Problem-agnostic Speech Representations from Multiple Self-supervised Tasks"
Pascual et al.
Paper: https://arxiv.org/abs/1904.03416
Code: https://github.com/santi-pdp/pase

Open Questions about Generative Adversarial Networks https://distill.pub/2019/gan-open-problems/ https://t.iss.one/ArtificialIntelligenceArticles

Repurposing CNNs - from images to sound:

Cool article from 2017 by Hershey et al. (https://arxiv.org/abs/1609.09430). Those nice folks took best of the best CNN for image recognition and repurposed them to identify audios. And, no wander, they succeeded.

What else looks like audio? Right, seismograms! Now I can't wait to implement ResNet for earthquake classification (which is already done btw

Step Change Improvement in Molecular Property Prediction with PotentialNet

Paper on a significant improvement in ability to predict molecular properties in drug design. #ML algorithms are getting better and better than classical methods.

Link: https://medium.com/@pandelab/step-change-improvement-in-molecular-property-prediction-with-potentialnet-f431ffa32a2c

#drugsdesign #biolearning #healthcare

Remember the black hole in the movie Interstellar? Turns out it was accurately modelled using Einstein's equations and 40000 lines of C++ code... and there's a full-on physics paper describing their process here: https://arxiv.org/pdf/1502.03808.pdf


#astrophysics #GravitationalLensing

Unsupervised learning: the curious pupil"
Unsupervised learning, a paradigm for creating artificial intelligence that learns about data without a particular task in mind: learning for the sake of learning.
Blog by Alexander Graves and Kelly Clancy, DeepMind: https://deepmind.com/blog/unsupervised-learning/
#artificialintelligence #deeplearning #unsupervisedlearning

Katie Bouman TED Talk
Katie Bouman is the postdoctoral fellow who led the development of the algorithm used to image a black hole.
TED Talk:

https://www.ted.com/talks/katie_bouman_what_does_a_black_hole_look_like

Fast Interactive Object Annotation with Curve-GCN"
Ling et al.
Paper: https://arxiv.org/pdf/1903.06874.pdf
Video: https://www.youtube.com/watch?v=ycD2BtO-QzU
#PyTorch Code: https://github.com/fidler-lab/curve-gcn
#ArtificialIntelligence #DeepLearning #MachineLearning

Best of arXiv.org for AI, Machine Learning, and Deep Learning – March 2019 #insidebigdata #BigDataAnalytics https://insidebigdata.com/2019/04/09/best-of-arxiv-org-for-ai-machine-learning-and-deep-learning-march-2019/

This one is a must read - the latest #ComputerVision #PatternRecognition https://deepai.org/publication/relational-action-forecasting

An overview of embedding models of entities and relationships for knowledge base completion

https://arxiv.org/abs/1703.08098

Liquid Splash Modeling with Neural Networks


https://ge.in.tum.de/download/2018-mlflip-um/2018-mlflip-um-talk.pdf

The Pros and Cons: Rank-aware Temporal Attention for Skill Determination in Long Videos

https://dimadamen.github.io/TheProsandCons/index.html

https://arxiv.org/pdf/1812.05538.pdf

Deep brain stimulation of the internal capsule enhances human cognitive control and prefrontal cortex function
https://www.nature.com/articles/s41467-019-09557-4

Heterogeneous Memory Enhanced Multimodal Attention Model for Video Question Answering. https://arxiv.org/abs/1904.04357