​​TOWARDS FEDERATED LEARNING AT SCALE: SYSTEM DESIGN

Google published how they do #FederatedLearning at scale on tens of millions of mobile phones. This is about training model on decentralized data.

ArXiV: https://arxiv.org/pdf/1902.01046.pdf

#Google #Privacy

​​Learning to Generalize from Sparse and Underspecified Rewards

Applying reinforcement learning to environments with sparse and underspecified rewards is an ongoing challenge, requiring generalization from limited feedback. Novel method that provides more refined feedback to the agent.

Link: https://ai.googleblog.com/2019/02/learning-to-generalize-from-sparse-and.html

#Google #RL

​​How 20th Century Fox uses ML to predict a movie audience

All modern blockbusters seem the same. They have common patterns of more exciting periods following less exciting, rotating emotional moments with action period. It is more about following well-known structure and template to make a well-boxing movie, than about director’s skill. No suprise, that #ML can be used to predict success of the movie by its trailer.

Link: https://cloud.google.com/blog/products/ai-machine-learning/how-20th-century-fox-uses-ml-to-predict-a-movie-audience

#DL #LAindustry #Google

Introducing Activation Atlases by #OpenAI

OpenAI in collaboration with #Google created activation atlases, a new technique for visualizing what interactions between neurons can represent.

Link: https://blog.openai.com/introducing-activation-atlases/
Direct demo link: https://distill.pub/2019/activation-atlas/app.html
Github: https://github.com/tensorflow/lucid/#activation-atlas-notebooks

#Google has open-sourced #FederatedLearning code

Step-by-step #tutorial showing how to perform Federated Learning using the same infrastructure Google
uses on 10s of millions of smartphones.

Link: https://medium.com/tensorflow/introducing-tensorflow-federated-a4147aa20041

Coconet: the ML model behind 20th of March Bach Doodle

Network trained to recreate Bach's music.

Link: https://magenta.tensorflow.org/coconet

#magenta #google #audiolearning

​​Reducing the Need for Labeled Data in Generative Adversarial Networks

How combination of self-supervision and semi-supervision can help learn from partially labeled data.

Link: https://ai.googleblog.com/2019/03/reducing-need-for-labeled-data-in.html

#GAN #DL #Google #supervisedvsunsupervised

​​Google announced the updated YouTube-8M dataset

Updated set now includes a subset with verified 5-s segment level labels, along with the 3rd Large-Scale Video Understanding Challenge and Workshop at #ICCV19.

Link: https://ai.googleblog.com/2019/06/announcing-youtube-8m-segments-dataset.html

#Google #YouTube #CV #DL #Video #dataset

​​XLNet: Generalized Autoregressive Pretraining for Language Understanding

Researchers at Google Brain and Carnegie Mellon introduce #XLNet, a pre-training algorithm for natural language processing systems. It helps NLP models (in this case, based on Transformer-XL) achieve state-of-the-art results in 18 diverse language-understanding tasks including question answering and sentiment analysis.

Article: https://towardsdatascience.com/what-is-xlnet-and-why-it-outperforms-bert-8d8fce710335
ArXiV: https://arxiv.org/pdf/1906.08237.pdf

#Google #GoogleBrain #CMU #NLP #SOTA #DL

​​Using Deep Learning to Inform Differential Diagnoses of Skin Diseases

Deep Learning System (DLS) for quicker and cheaper skin diseases detection. DLS showed accuracy across 26 skin conditions on par with U.S. board-certified dermatologists, when presented with identical information about a patient case (images and metadata). This is an amazing example of how technology can help fight notoriously high medical bills in the USA and make top-level care available and more affordable in all other the world.

Link: https://ai.googleblog.com/2019/09/using-deep-learning-to-inform.html?m=1
ArXiV: https://arxiv.org/abs/1909.05382

#Inception4 #Google

🔥 Tensorflow 2.0 release

Faster
TPU support
TensorFlow datasets

Change log: https://medium.com/tensorflow/tensorflow-2-0-is-now-available-57d706c2a9ab

#google #tensorflow #dl #tf

Looks like #google started fighting automatic #captcha recognition systems with adding noise to the images.

Large-Scale Multilingual Speech Recognition with a Streaming End-to-End Model

High-quality #speechrecognition systems require large amounts of data—yet many languages have little data available. Check out new research into an end-to-end system trained as a single model allowing for real-time multilingual speech recognition.

Link: https://ai.googleblog.com/2019/09/large-scale-multilingual-speech.html

#speech #audio #DL #Google

Simple comic on how #ML works from #Google

Make sure you save the link (or this message) to show it to people without great technical background for it is one of the best and clear explanations there is.

Link: https://cloud.google.com/products/ai/ml-comic-1/

#wheretostart #entrylevel #novice #explainingtochildren

​​🔥DeepMind’s AlphaStar beats top human players at strategy game StarCraft II

AlphaStar by Google’s DeepMind can now play StarCraft 2 so well that it places in the 99.8 percentile on the European server. In other words, way better than even great human players, achieving performance similar to gods of StarCraft.

Solution basically combines reinforcement learning with a quality-diversity algorithm, which is similar to an evolutionary algorithm.

What’s difficult about StarCraft and how is it different to recent #Go and #Chess AI solutions: even finding winning strategy (StarCraft is famouse to closeness to rock-scissors-paper, not-so-transitive game design, as chess and go), is not enough to win, since the result depends on execution on different macro and micro levels at different timescales.

How that is applicable in real world: basically, it is running logistics, manufacture, research with complex operations and different units.

Why this matters: it brings AI one step closer to running real business.

Blog post: https://deepmind.com/blog/article/AlphaStar-Grandmaster-level-in-StarCraft-II-using-multi-agent-reinforcement-learning
Nature: https://www.nature.com/articles/d41586-019-03298-6
ArXiV: https://arxiv.org/abs/1902.01724
Nontechnical video: https://www.youtube.com/watch?v=6eiErYh_FeY

#Google #GoogleAI #AlphaStar #Starcraft #Deepmind #nature #AlphaZero

🔥Human-like chatbots from Google: Towards a Human-like Open-Domain Chatbot.

TLDR: humanity is one huge step closer to a chat-bot, which can chat about anything and has great chance of success, passing #TuringTest

What does it mean: As an example, soon you will have to be extra-cautious chatting in #dating apps, because there will be more chat-bots, who can seem humane.
This also means that there will some positive and productive applications too: more sophisticated selling operators, on-demand psychological support, you name it.

It might be surprising, but #seq2seq still works. Over 5+ years of working on neural conversational models, general progress is a fine-tune of basic approach. It is a proof that much can be still discovered, along with room for new completely different approaches.

«Perplexity is all a chatbot needs ;)» (с) Quoc Le

Blog post: https://ai.googleblog.com/2020/01/towards-conversational-agent-that-can.html
Paper: https://arxiv.org/abs/2001.09977
Demo conversations: https://github.com/google-research/google-research/tree/master/meena

#NLP #NLU #ChatBots #google #googleai

​​AutoFlip: An Open Source Framework for Intelligent Video Reframing

Google released a tool for smart video cropping. Video cropping doesn't seem like a poblem until you release that object that should be in focus can be in different parts of picture. Now there is great attempt to provide one-click solution to cropping.

Interesting part: #AutoFlip is an application of #MediaPipe framework for building multimodal ML #pipelines.

Github: https://github.com/google/mediapipe/blob/master/mediapipe/docs/autoflip.md
MediaPipe: https://github.com/google/mediapipe/

#Google #GoogleAI #DL #CV

​​Racial Disparities in Automated Speech Recognition

To no surprise, speech recognition tools have #bias due to the lack of diversity in the datasets. Group of explorers addressed that issue and provided their’s research results as a paper and #reproducible research repo.

Project link: https://fairspeech.stanford.edu
Paper: https://www.pnas.org/cgi/doi/10.1073/pnas.1915768117
Github: https://github.com/stanford-policylab/asr-disparities

#speechrecognition #voice #audiolearning #dl #microsoft #google #apple #ibm #amazon

Lo-Fi Player

The team from the magenta project, that does research about deep learning and music powered by TensorFlow in Google, obviously, release a new fun project lofi-player powered by their open-source library magenta.js.

So it's basically a lo-fi music generator which popular genre on youtube streams and other kinds of stuff. You can customize the vibe on your manner and wish from sad to moody, slow to fast, etc.

It is based on their earlier work MusicVae to sample latent space of music and MelodyRNN to generate music sequences from different instruments. The project is not about new research, but to show what can do with an already done library in a creative way.

They also create a stream on youtube to listen lo-fi generated by that application and users in chat can together tune lo-fi player in stream :)

#magenta #lo-fi #music #google #tensorflow #fun