​Analogies from Word Vectors?
How analogous are these analogies?
https://towardsdatascience.com/analogies-from-word-vectors-77fe12f2de52?source=collection_home---4------3-----------------------

🔗 Analogies from Word Vectors?
How analogous are these analogies?

​Free-Lunch Saliency via Attention in Atari Agents
Dmitry Nikulin, Anastasia Ianina, Vladimir Aliev, Sergey Nikolenko

We propose a new approach to visualize saliency maps for deep neural network models and apply it to deep reinforcement learning agents trained on Atari environments. Our method adds an attention module that we call FLS (Free Lunch Saliency) to the feature extractor from an established baseline (Mnih et al., 2015). This addition results in a trainable model that can produce saliency maps, i.e., visualizations of the importance of different parts of the input for the agent's current decision making. We show experimentally that a network with an FLS module exhibits performance similar to the baseline (i.e., it is "free", with no performance cost) and can be used as a drop-in replacement for reinforcement learning agents. We also design another feature extractor that scores slightly lower but provides higher-fidelity visualizations. In addition to attained scores, we report saliency metrics evaluated on the Atari-HEAD dataset of human gameplay.
https://arxiv.org/abs/1908.02511

🔗 Free-Lunch Saliency via Attention in Atari Agents
We propose a new approach to visualize saliency maps for deep neural network models and apply it to deep reinforcement learning agents trained on Atari environments. Our method adds an attention module that we call FLS (Free Lunch Saliency) to the feature extractor from an established baseline (Mnih et al., 2015). This addition results in a trainable model that can produce saliency maps, i.e., visualizations of the importance of different parts of the input for the agent's current decision making. We show experimentally that a network with an FLS module exhibits performance similar to the baseline (i.e., it is "free", with no performance cost) and can be used as a drop-in replacement for reinforcement learning agents. We also design another feature extractor that scores slightly lower but provides higher-fidelity visualizations. In addition to attained scores, we report saliency metrics evaluated on the Atari-HEAD dataset of human gameplay.

​A Gentle Introduction to the Progressive Growing GAN
https://machinelearningmastery.com/introduction-to-progressive-growing-generative-adversarial-networks/

🔗 A Gentle Introduction to the Progressive Growing GAN
Progressive Growing GAN is an extension to the GAN training process that allows for the stable training of generator models that can output large high-quality images. It involves starting with a very small image and incrementally adding blocks of layers that increase the output size of the generator model and the input size of the …

​The data processing error in one of the most prominent fair machine learning datasets
ProPublica’s COMPAS Data Revisited

https://medium.com/@mbarenstein/the-data-processing-error-in-one-of-the-most-prominent-fair-machine-learning-datasets-4fa205daa3c4?source=topic_page---------5------------------1

🔗 The data processing error in one of the most prominent fair machine learning datasets
ProPublica’s COMPAS Data Revisited

​Геометрия машинного обучения. Разделяющие гиперплоскости или в чём геометрический смысл линейной комбинации?
https://habr.com/ru/post/324736/

🔗 Геометрия машинного обучения. Разделяющие гиперплоскости или в чём геометрический смысл линейной ком
Во многих алгоритмах машинного обучения, в том числе в нейронных сетях, нам постоянно приходится иметь дело со взвешенной суммой или, иначе, линейной комбинацией...

​Kaggle Santander Customer Transaction Prediction — Василий Рязанов

🔗 Kaggle Santander Customer Transaction Prediction — Василий Рязанов
Предсказание транзакций или поиск магии среди синтетических данных. Василий Рязанов рассказывает про опыт участия в Kaggle Santander Customer Transaction Prediction, где он заработал золотую медаль. Из видео вы сможете узнать: - Какие конкурсы Santander проводил на Kaggle - Странности, которые были замечены и попытки найти магию - Детали решения - Чего не хватило для победы Узнать о текущих соревнованиях можно на сайте http://mltrainings.ru/ Узнать о новых тренировках и видео можно из групп: ВКонтакте htt

​Audio samples related to Tacotron, an end-to-end speech synthesis system by Google.

https://google.github.io/tacotron/
https://arxiv.org/abs/1907.04448

🔗 Audio samples related to Tacotron, an end-to-end speech synthesis system by Google.

​How to Enter a Kaggle Competition (using Kernels) | Kaggle

🔗 How to Enter a Kaggle Competition (using Kernels) | Kaggle
Ever wanted to try out Kaggle competitions but weren't sure how to go about it? In this video Kaggle data scientist Rachael walks you through how to enter a competition to help you start your climb up the leaderboard! SUBSCRIBE: https://www.youtube.com/c/kaggle?sub_confirmation=1&utm_medium=youtube&utm_source=channel&utm_campaign=yt-sub About Kaggle: Kaggle is the world's largest community of data scientists. Join us to compete, collaborate, learn, and do your data science work. Kaggle's platform is the f

​Kaggle Days Dubai: Highlights | Kaggle Days

🔗 Kaggle Days Dubai: Highlights | Kaggle Days
Kaggle Days Dubai was held May 30 - April 1 2019 as a part of AI Everything. Participants came to meet, learn and code with Kaggle Grandmasters, and compete in a full-day offline competition. This edition is presented by LogicAI with sponsorship from Kaggle and Google Cloud. Kaggle Days are a global series of offline events for seasoned data scientists and Kagglers created by LogicAI and Kaggle. Follow Kaggle Days online: Visit the WEBSITE: https://kaggledays.com/ Like Kaggle Days on FACEBOOK: https://w

​More Data, More Sheets API
How the Google Sheets API uses magic (and code) to let you understand & visualize data.
https://medium.com/@yoyomade/more-data-more-sheets-api-b1e3ddfa0d1?source=topic_page---------6------------------1

🔗 More Data, More Sheets API
How the Google Sheets API uses magic (and code) to let you understand & visualize data.

​All about AI with Cheat-Sheets(+100 Cheat-sheets), Free Online Books, Courses, Videos and Lectures, Papers, Tutorials, Researchers, Websites, Datasets, Conferences, Frameworks, Tools
https://github.com/Niraj-Lunavat/Artificial-Intelligence

🔗 Niraj-Lunavat/Artificial-Intelligence
Awesome AI Learning with +100 AI Cheat-Sheets, Free online Books, Top Courses, Best Videos and Lectures, Papers, Tutorials, +99 Researchers, Premium Websites, +121 Datasets, Conferences, Frameworks...

​What I have Learned After Building A Successful AI PoC
I recently completed an AI PoC that has reached production and I wanted to share what I have learned on how to improve the chances
https://towardsdatascience.com/what-i-have-learned-after-building-a-successful-ai-poc-3bd24efea4e2?source=collection_home---4------1-----------------------

🔗 What I have Learned After Building A Successful AI PoC
I recently completed an AI PoC that has reached production and I wanted to share what I have learned on how to improve the chances of any…

​Proximal Policy Optimization Tutorial (Part 1: Actor-Critic Method)
Let’s code from scratch an RL football agent!
https://towardsdatascience.com/proximal-policy-optimization-tutorial-part-1-actor-critic-method-d53f9afffbf6?source=collection_home---4------0-----------------------

🔗 Proximal Policy Optimization Tutorial (Part 1: Actor-Critic Method)
Let’s code from scratch an RL football agent!

Flask and Deep Learning Keras/TensorFlow Web Services (13.1)

https://www.youtube.com/watch?v=H73m9XvKHug

🎥 Flask and Deep Learning Keras/TensorFlow Web Services (13.1)
👁 2 раз ⏳ 1061 сек.
To make use of your neural network in a website or other external application you will typically wrap the neural network in a RESTful application programming interface (API) through HTTP or HTTPS. This video shows how to use Flask to expose your Keras neural network as a RESTful API service.

Code for This Video:
https://github.com/jeffheaton/t81_558_deep_learning/blob/master/t81_558_class_13_01_flask.ipynb

Course Homepage: https://sites.wustl.edu/jeffheaton/t81-558/

Follow Me/Subscribe:
https://www.you

​MSURU: Large Scale E-commerce Image Classification With Weakly Supervised Search Data

https://research.fb.com/publications/msuru-large-scale-e-commerce-image-classification-with-weakly-supervised-search-data/

🔗 MSURU: Large Scale E-commerce Image Classification With Weakly Supervised Search Data
In this paper we present a deployed image recognition system used in a large scale commerce search engine, which we call MSURU. It is designed to process product images uploaded daily to Facebook Marketplace. Social commerce is a growing area within Facebook and understanding visual representations of product content is important for search and recommendation applications on Marketplace.

​Дайджест соревнований по анализу данных – Сергей Брянский

🔗 Дайджест соревнований по анализу данных – Сергей Брянский
Обзор новых соревнований по анализу данных от 10 августа 2019 года. Сергей Брянский рассказывает про актуальные соревнования, в которых можно принять участие. Календарь соревнований: http://mltrainings.ru/ Узнать о новых тренировках и видео можно из групп: ВКонтакте https://vk.com/mltrainings Facebook https://www.facebook.com/groups/1413405125598651/ Telegram https://t.me/mltrainings

​AutoDL 2019

https://autodl.chalearn.org/

🔗 AutoDL challenges

​RAPIDS: The platform inside and outside - Joshua Patterson | ODSC East 2019

🔗 RAPIDS: The platform inside and outside - Joshua Patterson | ODSC East 2019
Python has seen terrific progress as the data science language of choice. With the introduction of Pandas, users could interact with data in python in a way that fells intuitive. In addition, open-source packages such as Scikit-Learn have democratized and accelerated data science. RAPIDS seeks to have a similar impact on the Python data science community by accelerating data science with GPUs. RAPIDS is an open-source suite of tools for GPU data science. Launched in October, RAPIDS includes cuDF, a libr

​NVIDIA Clocks World’s Fastest BERT Training Time and Largest Transformer Based Model, Paving Path For Advanced Conversational AI

https://devblogs.nvidia.com/training-bert-with-gpus/

🔗 NVIDIA Clocks World’s Fastest BERT Training Time and Largest Transformer Based Model, Paving Path Fo
NVIDIA DGX SuperPOD trains BERT-Large in just 53 minutes, and trains GPT-2 8B, the largest Transformer Network Ever with 8.3Bn parameters Conversational AI is an essential building block of human interactions with intelligent machines and applications – from robots and cars, to home assistants and mobile apps. Getting computers to understand human languages, with all their …

​Видеокурс по нейронным сетям в Университете Шербрук
https://www.youtube.com/playlist?list=PL6Xpj9I5qXYEcOhn7TqghAJ6NAPrNmUBH - полный курс
Наш телеграм канал - tglink.me/ai_machinelearning_big_data

🔗 Neural networks class - Université de Sherbrooke - YouTube
These are the videos I use to teach my Neural networks class at Université de Sherbrooke. The videos, along with the slides and research paper references, ar...


🎥 Neural networks [1.1] : Feedforward neural network - artificial neuron
👁 1 раз ⏳ 471 сек.


🎥 Neural networks [1.2] : Feedforward neural network - activation function
👁 1 раз ⏳ 356 сек.


🎥 Neural networks [1.3] : Feedforward neural network - capacity of single neuron
👁 1 раз ⏳ 485 сек.


🎥 Neural networks [1.4] : Feedforward neural network - multilayer neural network
👁 1 раз ⏳ 791 сек.