Feature-wise transformations
New review article on feature-wise transformations (like FiLM) in the highly visual and interactive Distill journal. With Vincent Dumoulin, Ethan Perez, Nathan Schucher, Florian Strub, Harm de Vries, Aaron Courville and Yoshua Bengio. Many institutions involved. A very useful building block for learning conditioning computation (especially visual computation).
https://distill.pub/2018/feature-wise-transformations/
@ArtificialIntelligenceArticles
New review article on feature-wise transformations (like FiLM) in the highly visual and interactive Distill journal. With Vincent Dumoulin, Ethan Perez, Nathan Schucher, Florian Strub, Harm de Vries, Aaron Courville and Yoshua Bengio. Many institutions involved. A very useful building block for learning conditioning computation (especially visual computation).
https://distill.pub/2018/feature-wise-transformations/
@ArtificialIntelligenceArticles
Distill
Feature-wise transformations
A simple and surprisingly effective family of conditioning mechanisms.
DeepMind papers at ICML 2018
https://deepmind.com/blog/deepmind-papers-icml-2018/
@ArtificialIntelligenceArticles
https://deepmind.com/blog/deepmind-papers-icml-2018/
@ArtificialIntelligenceArticles
DeepMind
DeepMind papers at ICML 2018 | DeepMind
The 2018 International Conference on Machine Learning will take place in Stockholm, Sweden from 10-15 July. Here you will find a schedule of DeepMind presentations at ICML.
ICML 2018 Tutorial sessions
International Conference on Machine Learning @icmlconf (ICML 2018) a premier annual Machine Learning event is going Live Stream at
ICML 2018 Tutorial sessions (Live Stream )
https://www.facebook.com/icml.imls/videos/428757614305426/
@ArtificialIntelligenceArticles
International Conference on Machine Learning @icmlconf (ICML 2018) a premier annual Machine Learning event is going Live Stream at
ICML 2018 Tutorial sessions (Live Stream )
https://www.facebook.com/icml.imls/videos/428757614305426/
@ArtificialIntelligenceArticles
Facebook
International Conference on Machine Learning
Welcome back to the ICML 2018 Tutorial sessions. This tutorial Optimization Perspectives on Learning to Control will survey the foundations required to build machine learning systems that reliably...
Toward theoretical understanding of deep learning
https://unsupervised.cs.princeton.edu/deeplearningtutorial.html
video :
https://www.facebook.com/icml.imls/videos/428562880991566/
@ArtificialIntelligenceArticles
https://unsupervised.cs.princeton.edu/deeplearningtutorial.html
video :
https://www.facebook.com/icml.imls/videos/428562880991566/
@ArtificialIntelligenceArticles
unsupervised.cs.princeton.edu
Toward Theoretical Understanding of Deep Learning: Slides and
Bibliography
Bibliography
Material to accompany ICMLR'18
tutorial.
tutorial.
Facebook Research at ICML 2018
https://research.fb.com/facebook-research-at-icml-2018/
@ArtificialIntelligenceArticles
https://research.fb.com/facebook-research-at-icml-2018/
@ArtificialIntelligenceArticles
Optimization Perspectives on Learning to Control #ICML2018 #Tutorial
https://people.eecs.berkeley.edu/~brecht/l2c-icml2018/
@ArtificialIntelligenceArticles
https://people.eecs.berkeley.edu/~brecht/l2c-icml2018/
@ArtificialIntelligenceArticles
First human body scan performed.
First 3D colour X-ray of a human using CERN technology https://home.cern/about/updates/2018/07/first-3d-colour-x-ray-human-using-cern-technology
First 3D colour X-ray of a human using CERN technology https://home.cern/about/updates/2018/07/first-3d-colour-x-ray-human-using-cern-technology
Inability of convolutional #neuralnetworks (CNNs) to transform spatial representations between two different types: coordinates in (i, j) Cartesian space and coordinates in one-hot pixel space.
https://eng.uber.com/coordconv/
paper: https://arxiv.org/abs/1807.03247
https://eng.uber.com/coordconv/
paper: https://arxiv.org/abs/1807.03247
Uber Blog
An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution | Uber Blog
As powerful and widespread as convolutional neural networks are in deep learning, AI Labs’ latest research reveals both an underappreciated failing and a simple fix.
AI Can Now Fix Your Grainy Photos by Only Looking at Grainy Photos https://goo.gl/3C6ZpS
paper: https://arxiv.org/abs/1803.04189 @ArtificialIntelligenceArticles
paper: https://arxiv.org/abs/1803.04189 @ArtificialIntelligenceArticles
The matrix calculus you need for #deeplearning
https://explained.ai/matrix-calculus/index.html or
paper : arxiv.org/abs/1802.01528
@ArtificialIntelligenceArticles
https://explained.ai/matrix-calculus/index.html or
paper : arxiv.org/abs/1802.01528
@ArtificialIntelligenceArticles
Capsules for Object Segmentation.
Paper: https://arxiv.org/abs/1804.04241
Code: https://goo.gl/x1qh4f Project page: https://goo.gl/MYzWfY
Paper: https://arxiv.org/abs/1804.04241
Code: https://goo.gl/x1qh4f Project page: https://goo.gl/MYzWfY
Live video of the ICML Test Of Time Award talk by FAIRies Ronan Collobert and Jason Weston for their 2008 ICML paper that pioneered deep learning approaches to natural language processing.
https://www.facebook.com/icml.imls/videos/432124137302107/
@ArtificialIntelligenceArticles
https://www.facebook.com/icml.imls/videos/432124137302107/
@ArtificialIntelligenceArticles
Facebook
International Conference on Machine Learning
Welcome to ICML 2018 live from Stockholm. In this session, Test Of Time Award is presented to Ronan Collobert and Jason Weston for A Unified Architecture for Natural Language Processing: Deep Neural...
Universal Transformers
"achieves a new state of the art on the bAbI linguistic reasoning task"
paper: https://arxiv.org/abs/1807.03819
source code: https://github.com/tensorflow/tensor2tensor
#GoogleBrain
@ArtificialIntelligenceArticles
"achieves a new state of the art on the bAbI linguistic reasoning task"
paper: https://arxiv.org/abs/1807.03819
source code: https://github.com/tensorflow/tensor2tensor
#GoogleBrain
@ArtificialIntelligenceArticles
GitHub
GitHub - tensorflow/tensor2tensor: Library of deep learning models and datasets designed to make deep learning more accessible…
Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research. - tensorflow/tensor2tensor
مقاله ی جدیدی از پروفسور جفری هینتون منتشر شده در تاریخ 12 Jul
paper : https://arxiv.org/abs/1807.04587v1 @ArtificialIntelligenceArticles
paper : https://arxiv.org/abs/1807.04587v1 @ArtificialIntelligenceArticles
بهترین مقالات arXiv مرتبط با هوش مصنوعی یادگیری ماشین و یادگیری عمیق
Best of arXiv.org for AI, Machine Learning, and Deep Learning https://goo.gl/bZf8U7 @ArtificialIntelligenceArticles
Best of arXiv.org for AI, Machine Learning, and Deep Learning https://goo.gl/bZf8U7 @ArtificialIntelligenceArticles
Winning solutions of past #kaggle competitions in one place
https://www.kaggle.com/sudalairajkumar/winning-solutions-of-kaggle-competitions @ArtificialIntelligenceArticles
https://www.kaggle.com/sudalairajkumar/winning-solutions-of-kaggle-competitions @ArtificialIntelligenceArticles
Seedbank, a place to discover interactive ML examples which you can run from your browser, no set-up required.
Each example can be edited, extended, and adapted into your own project.
Blog post by Michael Tyka: https://medium.com/tensorflow/seedbank-discover-machine-learning-examples-2ff894542b57?linkId=54191031
@ArtificialIntelligenceArticles
Each example can be edited, extended, and adapted into your own project.
Blog post by Michael Tyka: https://medium.com/tensorflow/seedbank-discover-machine-learning-examples-2ff894542b57?linkId=54191031
@ArtificialIntelligenceArticles
Medium
Seedbank — discover machine learning examples
By Michael Tyka, Artists + Machine Intelligence
Very detailed notes of #ICML2018 conference. 55 pages from David Abel. https://goo.gl/kAsFTC @ArtificialIntelligenceArticles
Learning World Models: The Next Step Towards AI. Invited talk by Yann LeCun
https://www.facebook.com/ijcaiecai18/videos/2184672041673770/ @ArtificialIntelligenceArticles
https://www.facebook.com/ijcaiecai18/videos/2184672041673770/ @ArtificialIntelligenceArticles