If an eye doctor looked at a retinal photo, the chance of getting gender correct would be 50-50.
But deep learning training led to an AUC of 0.97 https://www.nature.com/articles/s41551-018-0195-0

Attention and Memory in Deep Learning
https://www.youtube.com/watch?v=Q57rzaHHO0k @ArtificialIntelligenceArticles

Yandex Data School: Course in Natural Language Processing

Github Course by Russian Search Giant
* lectures
* seminars
* home assignments

https://github.com/yandexdataschool/nlp_course

The Physics of Brain Network Structure. #BigData #Analytics #DeepLearning https://arxiv.org/abs/1809.06441 @ArtificialIntelligenceArticles

Machine Learning For Jet Physics: New, Or Just Cool, Ideas https://www.science20.com/tommaso_dorigo/machine_learning_for_jet_physics_new_or_just_cool_ideas-235272 @ArtificialIntelligenceArticles

Generating a Training Dataset for Land Cover Classification to Advance Global Development. https://arxiv.org/abs/1811.07998

SQuAD2.0: The Stanford Question Answering Dataset

Leaderboard: https://rajpurkar.github.io/SQuAD-explorer/

#artificialintelligence #bigdata #deeplearning #machinelearning #NaturalLanguageProcessing

Predicting Diabetes Disease Evolution Using Financial Records and Recurrent Neural Networks https://arxiv.org/abs/1811.09350

DeepMasterPrints: Generating MasterPrints for Dictionary Attacks via Latent Variable Evolution"

Bontrager et al.: https://arxiv.org/abs/1705.07386

Do Better ImageNet Models Transfer Better?

By Simon Kornblith, Jonathon Shlens, Quoc V. Le: https://arxiv.org/abs/1805.08974

#ComputerVision #PatternRecognition #MachineLearning

Explainable cardiac pathology classification on cine MRI with motion characterization https://arxiv.org/abs/1811.03433

Measuring the Effects of Data Parallelism on Neural Network Training

Important paper from Google on large batch optimization. They do impressively careful experiments measuring # iterations needed to achieve target validation error at various batch sizes. The main "surprise" is the lack of surprises.


https://arxiv.org/abs/1811.03600 @ArtificialIntelligenceArticles

Statistical physics of liquid brains

Paper by Jordi Pinero and Ricard Sole: https://www.biorxiv.org/content/biorxiv/early/2018/11/26/478412.full.pdf

#Brains #Evolution #Physics

Visualizing the Loss Landscape of Neural Nets

PyTorch code by Tom Goldstein: https://github.com/tomgoldstein/loss-landscape

#pytorch #neuralnetworks #machinelearning #deeplearning

A primer on deep learning in genomics https://www.nature.com/articles/s41588-018-0295-5 @ArtificialIntelligenceArticles

Dive into Deep Learning

Jupyter Notebooks, PDF, and website, all generated from one source.

By Zhang et al.: https://www.diveintodeeplearning.org/

#machinelearning #deeplearning #artificialintellige

The same #MachineLearning #courses used to train engineers at #Amazon are now available to all developers through #AWS. More than 30 self-service, self-paced digital courses with more than 45 hours of courses, videos, and labs are available at no charge!


https://aws.amazon.com/training/learning-paths/machine-learning/


https://aws.amazon.com/blogs/machine-learning/amazons-own-machine-learning-university-now-available-to-all-developers/

https://t.iss.one/ArtificialIntelligenceArticles

TorchCraftAI

A bot platform for machine learning research on StarCraft: Brood War

GitHub: https://github.com/TorchCraft/TorchCraftAI

Deep Learning and Density Functional Theory. https://arxiv.org/abs/1811.08928

3D Hair Synthesis Using Volumetric Variational Autoencoders https://linjieluo.com/publications/3d-hair-synthesis-using-volumetric-variational-autoencoders/ @ArtificialIntelligenceArticles