Команда Google DeepMind представила новый ИИ, способный самостоятельно учиться выполнять задачи
https://tproger.ru/news/deepmind-new-algorithm/

Port of Single Shot MultiBox Detector to Keras (SSD)
https://arxiv.org/pdf/1512.02325v3.pdf
https://github.com/rykov8/ssd_keras
#keras #ssd

Курс по Tensorflow от BigDataUniversity (free)
https://bigdatauniversity.com/courses/deep-learning-tensorflow/

ML0120EN
COURSE LEVEL: Advanced
TIME TO COMPLETE: 10 Hours

#tensorflow

cnn-benchmarks
Benchmarks for popular convolutional neural network models on CPU and different GPUs, with and without cuDNN.

https://github.com/jcjohnson/cnn-benchmarks

Презентация
Deep learning for audio-based music recommendation

We are delighted to republish slides on Deep learning for audio-based
music recommendation by Sander Dieleman.

Sander is a Research Scientist at DeepMind, and was previously involved in scaling up content-based music recommendation at Spotify. Sander is a PhD student at Ghent University.

The presentation covers deep content-based music recommendation approach that is an alternative solution to widely adopted collaborative filtering.

Initially slides were presented at Workshop on Deep Learning for Recommender Systems in Bosoton in September 2016.

http://www.russia.ai/single-post/2016/12/01/Deep-learning-for-audio-based-music-recommendation

Видео доклада про практики ЖЦ систем машинного обучения

http://ailev.livejournal.com/1316740.html?utm_source=vksharing&utm_medium=social

Implementing a CNN for Human Activity Recognition in Tensorflow

Histogram of Oriented Gradients | Learn OpenCV
www.learnopencv.com/histogram-of-oriented-gradients/

Сооснователь Maps.me Юрий Мельничек выпустил приложение для обработки фона фотографий с помощью нейросетей

Сегментацию делает CNN на GPU (на старых телефонах на CPU)
Работает по похожему принципу http://xiaoyongshen.me/webpage_portrait/papers/portrait_eg16.pdf

Обсуждение в telegram https://telegram.me/joinchat/ABI4pz6rz2iVzWUzaVqpmA