SLIDES
What can Statistical Machine Translation teach Neural Text Generation about Optimization

Graham Neubig

@ NAACL Workshop on Methods for Optimizing and Evaluating Neural Language Generation6/6/2019
https://www.phontron.com/slides/neubig19neuralgen.pdf

Course 3 of the deeplearning.ai TensorFlow Specialization is now available on Coursera! You’ll learn how to build natural language processing systems using TensorFlow. Enroll in the Specialization for $49/month or audit for free: https://www.coursera.org/specializations/tensorflow-in-practice

Losing Confidence in Quality: Unspoken Evolution of Computer Vision Services," Cummaudo et al.: https://arxiv.org/abs/1906.07328

A new research paper by Geoffry E.Hinton , Adam R. Kosiorek, Sara Sabour, Yee Whye Teh

a new version of capsule networks, called Stacked Capsule Autoencoders


https://arxiv.org/abs/1906.06818

https://t.iss.one/ArtificialIntelligenceArticles

Stanford Machine Learning Class Notes (CS229)
BY TANUJIT CHAKRABORTY

Download Link: https://www.ctanujit.org/uploads/2/5/3/9/25393293/machine_learning_notes__cs229_.pdf

Published in CVPR 2019: Stanford researchers address weakly-supervised action alignment and segmentation in videos, where only the order of occurring actions is available during training.

Their model outperforms the current state-of-the-art

https://www.profillic.com/paper/arxiv:1901.02598

Open source PyRobot, a lightweight, high-level interface that lets AI researchers get up and running with robotics experiments in just hours. No specialized robotics expertise needed
https://ai.facebook.com/blog/open-sourcing-pyrobot-to-accelerate-ai-robotics-research/

https://www.google.com/amp/s/amp.cnn.com/cnn/2019/04/14/africa/google-ai-center-accra-intl/index.html Google Google AI

Kaggle dataset usability ratings on 17000+ public datasets

Here: https://www.kaggle.com/datasets

#ArtificialIntelligence #DeepLearning #MachineLearning

Yann lecun : Great interview with Peter Shor.


https://blogs.scientificamerican.com/cross-check/quantum-computing-for-english-majors/

Self-Supervised Learning

Tutorial by Andrew Zisserman: https://project.inria.fr/paiss/files/2018/07/zisserman-self-supervised.pdf

#CVPR #DeepLearning #SelfSupervisedLearning

Language as an Abstraction for Hierarchical Deep Reinforcement Learning

Jiang et al.: https://arxiv.org/abs/1906.07343

#reinforcementlearning #language #machinelearning

The researchers have constructed a ghostwriter program which utilizes a [Siamese neural network](https://www.cs.cmu.edu/~rsalakhu/papers/oneshot1.pdf). This process can distinguish the writing styles of two texts. Over time the network is trained using voluminous amounts of data to learn from representations of writing styles (in this case, 130,000 essays were examined from 10,000 students). These are the compared by the program. Siamese neural networks are also being used for recognizing handwritten checks, automatic detection of faces in camera images, and matching queries with indexed documents.

Read more: [https://www.digitaljournal.com/tech-and-science/technology/ai-can-now-catch-90-percent-of-essay-paper-cheats/article/551126#ixzz5qTZyK73D](https://www.digitaljournal.com/tech-and-science/technology/ai-can-now-catch-90-percent-of-essay-paper-cheats/article/551126#ixzz5qTZyK73D)

EasyGen, a visual programming language for text data pipelines for neural nets.
By Mark Riedl.
Colab: https://drive.google.com/open?id=1XNiOuNtMnItl5CPGvRjEvj9C78nDuvXj
Github: https://github.com/markriedl/easygen
#ArtificialIntelligence #MachineLearning #NeuralNetworks

Neurobiologists train artificial neural networks to map the brain
https://bit.do/eVNef

#cellularmorpoholopyneuralnetworks #unsupervisedlearning
#analyzinglargedatasets #CNN #AI

The human brain consists of about 86 billion nerve cells and about as many glial cells. In addition, there are about 100 trillion connections between the nerve cells alone. While mapping all the connections of a human brain remains out of reach, scientists have started to address the problem on a smaller scale. Through the development of serial block-face scanning electron microscopy, all cells and connections of a particular brain area can now be automatically surveyed and displayed in a three-dimensional image.

“It can take several months to survey a 0.3 mm3 piece of brain under an electron microscope. Depending on the size of the brain, this seems like a lot of time for a tiny piece. But even this contains thousands of cells. Such a data set would also require almost 100 terabytes of storage space. However, it is not the collection and storage but rather the data analysis that is the difficult part."

Machine Learning Crash Course with TensorFlow APIs by Google
Free Course

LInk: https://developers.google.com/machine-learning/crash-course/

Build Realistic Human Speech Animations with the New VOCA Model and 4D Face Dataset
https://medium.com/syncedreview/build-realistic-human-speech-animations-with-the-new-voca-model-and-4d-face-dataset-d26ebca37bb7

ICYMI: Facebook researchers open sourced PyRobot, a lightweight, high-level interface that lets AI researchers get up and running with robotics experiments in just hours. No specialized robotics expertise needed!

https://www.profillic.com/paper/arxiv:1906.08236

Distributed Deep Learning Pipelines with PySpark and Keras
https://towardsdatascience.com/distributed-deep-learning-pipelines-with-pyspark-and-keras-a3a1c22b9239

The Functional Neural Process
Louizos et al.: https://arxiv.org/abs/1906.08324
#ArtificialIntelligence #Bayesian #MachineLearning

A scientist at Google Brain devised a way for a machine-learning system to teach itself about how the world works. His name is Ian Goodfellow, and he was one of our 35 Innovators Under 35 in 2017. This year's list comes out on June 25. Stay tuned for the 35 inventors, entrepreneurs, visionaries, humanitarians, and pioneers who will shape tomorrow's technology.
https://www.technologyreview.com/lists/innovators-under-35/2017/inventor/ian-goodfellow/