Using #AI To Analyze Video As Imagery: The Impact Of Sampling Rate
https://buff.ly/2IBFRO8
#ArtificialIntelligence #MachineLearning #DeepLearning #robotics

✴️ @AI_Python_EN

Bayesian Optimization with Binary Auxiliary Information
https://deepai.org/publication/bayesian-optimization-with-binary-auxiliary-information … by Yehong Zhang et al.
#ReinforcementLearning #Hyperparameter

✴️ @AI_Python_EN

Best Paper Finalist (top 1% of accepted papers) Check it out!
http://openaccess.thecvf.com/content_CVPR_2019/html/Ribera_Locating_Objects_Without_Bounding_Boxes_CVPR_2019_paper.html

Comparison of different #MachineLearning approaches for neuroimaging data
Main take-aways - prediction accuracy increased once N ≥ 400
- Substantial effect of pipeline on accuracies: Is this the new p-hacking?
https://buff.ly/2NdaJcv

✴️ @AI_Python_EN

Theory of the Frequency Principle for General Deep Neural Networks.
http://arxiv.org/abs/1906.09235
✴️ @AI_Python_EN

Artificial Intelligence can write creative & convincingly human-like captions for any image. Great work by IBM Research at #cvpr2019 In order to ensure the generated captions did not sound too unnatural, the work employed conditional GAN training Read
https://arxiv.org/pdf/1805.00063.pdf

✴️ @AI_Python_EN

#Machineearning for Everyone
http://bit.ly/2RvRRnj

#AI #ML #DataScience #Algorithms

✴️ @AI_Python_EN

A Gentle Introduction to Upsampling and Transpose Convolution Layers for GANs

Generative Adversarial Networks, or GANs, are an architecture for training generative models, such as deep convolutional neural networks for generating images

✴️ @AI_Python_EN

The #xtensor article series continues! Learn everything about xtensor constructors and initializer lists in

https://medium.com/@johan.mabille/how-we-wrote-xtensor-3-n-the-constructors-65a177260638


✴️ @AI_Python_EN

This is incredible. This paper from MIT Computer Science & Artificial Intelligence Lab presented at #cvpr2019 shows how to reconstruct a face from speech patterns.
https://speech2face.github.io

✴️ @AI_Python_EN

Using #DeepLearning to produce an #AutonomousSystem for detecting traffic signs on Google Street View images. The system could help to monitor street signs and identify those in need of replacement or repair.
Read: http://ow.ly/WW4630oZzJu

https://doi.org/10.1016/j.compenvurbsys.2019.101350

✴️ @AI_Python_EN

All the datasets (there are a lot) released at #cvpr2019 are now indexed in
http://visualdata.io . Check them out!
#computervision #machinelearning #dataset

✴️ @AI_Python_EN

summary of interesting optimization papers from ICLR 2019:
part I :
https://medium.com/@yaroslavvb/iclr-optimization-papers-i-fluctuation-dissipation-relations-for-sgd-a638ad9964cc
part II
https://medium.com/@yaroslavvb/iclr-optimization-papers-ii-44f03b98dc5f?postPublishedType=repub
part III
https://medium.com/@yaroslavvb/iclr-optimization-papers-iii-1e1edc050ba6

✴️ @AI_Python_EN

decentralized decision making is out on arxiv: "Reinforcement Learning with Competitive Ensembles of Information-Constrained Primitives".
Link: https://arxiv.org/abs/1906.10667 .

✴️ @AI_Python_EN

Look at this amazing collection of resources for teaching reproducible research to university students! 👨‍🎓.
https://guides.lib.uw.edu/research/reproducibility/teaching

✴️ @AI_Python_EN

eleased additional "image-wise" models from our paper "Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening". These models act on images individually making them well suited for transfer learning.
https://github.com/nyukat/breast_cancer_classifier

✴️ @AI_Python_EN

Stunning big picture #infographic depicting how #MachineLearning works, its relationship to #AI, and where companies are putting it to work.

✴️ @AI_Python_EN

talk at PyData Amsterdam 2019 on our low-to-high resolution project is out! If you missed his talk at PyData Amsterdam or in general if you're interested in image super resolution, check out his video and also of course our Github repo for more information. #deeplearning #machinelearning #AxelSpringerAI

▶️ YouTube Video: https://lnkd.in/d6YHaFS
🔤 Code: https://lnkd.in/dkJUaQe

✴️ @AI_Python_EN

Quick links for all things #R and #Python:

1. Overview of using python with RStudio: https://lnkd.in/d5NkJAt
2. Python & #shiny: https://lnkd.in/dVfkE6b
3. Python & #rmarkdown: https://lnkd.in/dXpSd7i
4. Python with #plumber: https://lnkd.in/dn2pEAQ

For a central location to publish all of your team's data products (R artifacts, R & python mixed assets, and #jupyternotebooks), check out RStudio Connect: https://lnkd.in/dXW7iPG

✴️ @AI_Python_EN

Human in the Loop: Deep Learning without Wasteful Labelling

Kirsch et al.: https://lnkd.in/eP323W3

Code: https://lnkd.in/e7-wbxD

#activelearning #deeplearning #informationtheory
#machinelearning

✴️ @AI_Python_EN

The same statistical or machine learning method can be programmed (implemented) in different ways, and this can have an impact on the results. (I'm not referring to programing errors.)

Moreover, the initial start seed can strongly affect a routine - change the start seed and the results may vary substantially.

So, the same method programmed the same way may give different results on the same data if you change the start seed.

Most (hopefully all) statisticians are aware of this, but I suspect most users (e.g., decision makers) are not. "AI" is not immune to this.

✴️ @AI_Python_EN