tensorflow/cleverhans: a Python library to benchmark machine learning systems' vulnerability to adversarial examples
https://github.com/tensorflow/cleverhans
https://github.com/tensorflow/cleverhans
GitHub
GitHub - cleverhans-lab/cleverhans: An adversarial example library for constructing attacks, building defenses, and benchmarking…
An adversarial example library for constructing attacks, building defenses, and benchmarking both - cleverhans-lab/cleverhans
Code for "On the Effects of Batch and Weight Normalization in Generative Adversarial Networks"
https://github.com/stormraiser/GAN-weight-norm
https://github.com/stormraiser/GAN-weight-norm
GitHub
stormraiser/GAN-weight-norm
Code for "On the Effects of Batch and Weight Normalization in Generative Adversarial Networks" - stormraiser/GAN-weight-norm
Learning Musical Style and Generating Musical Performances using LSTMs
http://imanmalik.com/cs/2017/06/05/neural-style.html
http://imanmalik.com/cs/2017/06/05/neural-style.html
Finding bad flamingo drawings with recurrent neural networks
http://colinmorris.github.io/blog/bad_flamingos
http://colinmorris.github.io/blog/bad_flamingos
colinmorris.github.io
Finding bad flamingo drawings with recurrent neural networks
Performing anomaly detection using a recurrent neural network to find the ugliest flamingos in a dataset of sketches.
Interactive tutorial: generative adversarial networks for beginners, with TensorFlow
https://www.oreilly.com/learning/generative-adversarial-networks-for-beginners
https://www.oreilly.com/learning/generative-adversarial-networks-for-beginners
O’Reilly Media
Generative Adversarial Networks for beginners
Build a neural network that learns to generate handwritten digits.
Gradient Boosting – the coolest kid on the machine learning block
https://www.displayr.com/gradient-boosting-the-coolest-kid-on-the-machine-learning-block/
https://www.displayr.com/gradient-boosting-the-coolest-kid-on-the-machine-learning-block/
Displayr
Gradient Boosting Explained - The Coolest Kid on The Machine Learning Block
Gradient boosting is attracting attention for its prediction speed & accuracy, especially with large & complex data. Learn about gradient boosting here.
Portraits of Imaginary people. GANs at 4000x4000 pixel resolution.
http://mtyka.github.io/machine/learning/2017/06/06/highres-gan-faces.html
http://mtyka.github.io/machine/learning/2017/06/06/highres-gan-faces.html
mtyka.github.io
Work in progress: Portraits of Imaginary People
For a while now I’ve been experimenting with ways to use generative neural nets to make portraits. Early experiments were based on deepdream-like approaches ...
ML notes: Why the log-likelihood?
https://blog.metaflow.fr/ml-notes-why-the-log-likelihood-24f7b6c40f83
https://blog.metaflow.fr/ml-notes-why-the-log-likelihood-24f7b6c40f83
Medium
ML notes: Why the log-likelihood?
Disclaimer:
Variational Inference and Deep Learning: An Intuitive Introduction (by Alex Lamb)
https://www.youtube.com/watch?v=h0UE8FzdE8U
https://www.youtube.com/watch?v=h0UE8FzdE8U
YouTube
Variational Inference and Deep Learning: An Intuitive Introduction
A lecture introducing Variational Inference and Deep Learning. Adapted from a lecture I gave for Aaron Courville's Deep Learning course (IFT 6266).
Доклады с source{d} митапа про ML на исходном коде
https://www.youtube.com/playlist?list=PL5Ld68ole7j3iQFUSB3fR9122dHCUWXsy
https://www.youtube.com/playlist?list=PL5Ld68ole7j3iQFUSB3fR9122dHCUWXsy
YouTube
source{d} tech talks - Machine Learning 2017 - YouTube
Predicting Football Results With Statistical Modelling
https://dashee87.github.io/football/python/predicting-football-results-with-statistical-modelling/
https://dashee87.github.io/football/python/predicting-football-results-with-statistical-modelling/
dashee87.github.io
Predicting Football Results With Statistical Modelling
Combining the world’s most popular sport with everyone’s favourite discrete probability distribution, this post predicts football matches using the Poisson distribution.
Градиентный бустинг
https://alexanderdyakonov.wordpress.com/2017/06/09/%D0%B3%D1%80%D0%B0%D0%B4%D0%B8%D0%B5%D0%BD%D1%82%D0%BD%D1%8B%D0%B9-%D0%B1%D1%83%D1%81%D1%82%D0%B8%D0%BD%D0%B3/#more-5246
https://alexanderdyakonov.wordpress.com/2017/06/09/%D0%B3%D1%80%D0%B0%D0%B4%D0%B8%D0%B5%D0%BD%D1%82%D0%BD%D1%8B%D0%B9-%D0%B1%D1%83%D1%81%D1%82%D0%B8%D0%BD%D0%B3/#more-5246
Анализ малых данных
Градиентный бустинг
Пост про градиентный бустинг (Gradient Boosting), но не совсем обычный. Вместо текста прикрепляю pdf. Вопрос к читателям блога: будет ли полезно, если я подготовлю книжку в таком стиле по основным …
Playing a toy poker game with Reinforcement Learning
http://willtipton.com/coding/poker/2017/06/06/shove-fold-with-reinforcement-learning.html
http://willtipton.com/coding/poker/2017/06/06/shove-fold-with-reinforcement-learning.html
Willtipton
Playing a toy poker game with Reinforcement Learning
Coding and poker, mostly.
A DJ Khaled themed object recognizer app using inception v3, iOS CoreML and Vision
https://github.com/G2Jose/CoreMLVisionDJKhaled
https://github.com/G2Jose/CoreMLVisionDJKhaled
GitHub
G2Jose/CoreMLVisionDJKhaled
CoreMLVisionDJKhaled - Check if an object seen by your camera is a lion. Uses iOS CoreML, Vision APIs
MNIST FOR ML BEGINNERS: THE BAYESIAN WAY
https://alpha-i.co/blog/MNIST-for-ML-beginners-The-Bayesian-Way.html
https://alpha-i.co/blog/MNIST-for-ML-beginners-The-Bayesian-Way.html
CortexNet: a robust predictive deep neural network trained on videos
https://engineering.purdue.edu/elab/CortexNet/
https://engineering.purdue.edu/elab/CortexNet/
Google releases imagenet pre-trained mobilenet (faster/more-accurate than alexnet) models
https://research.googleblog.com/2017/06/mobilenets-open-source-models-for.html
https://research.googleblog.com/2017/06/mobilenets-open-source-models-for.html
research.google
MobileNets: Open-Source Models for Efficient On-Device Vision
Posted by Andrew G. Howard, Senior Software Engineer and Menglong Zhu, Software Engineer(Cross-posted on the Google Open Source Blog)Deep learning ...
Dropout — метод решения проблемы переобучения в нейронных сетях
https://habrahabr.ru/company/wunderfund/blog/330814/
https://habrahabr.ru/company/wunderfund/blog/330814/
Хабр
Dropout — метод решения проблемы переобучения в нейронных сетях
Переобучение (overfitting) — одна из проблем глубоких нейронных сетей (Deep Neural Networks, DNN), состоящая в следующем: модель хорошо объясняет только пример...
Один из самых значимых результатов в AI Safety (и не только) за последнее время, полученный при коллаборации исследователей из OpenAI и Deepmind.
https://blog.openai.com/deep-reinforcement-learning-from-human-preferences/
https://blog.openai.com/deep-reinforcement-learning-from-human-preferences/
OpenAI
Learning from Human Preferences
One step towards building safe AI systems is to remove the need for humans to
write goal functions, since using a simple proxy for a complex goal, or getting
the complex goal a bit wrong, can lead to undesirable and even dangerous
behavior [https://arxiv…
write goal functions, since using a simple proxy for a complex goal, or getting
the complex goal a bit wrong, can lead to undesirable and even dangerous
behavior [https://arxiv…