​Data Science vs Machine Learning – What’s The Difference? | Data Science Course | Edureka

🔗 Data Science vs Machine Learning – What’s The Difference? | Data Science Course | Edureka
**Python Data Science Training: https://www.edureka.co/python ** In this video on Data Science vs Machine Learning, we’ll be discussing the importance of Dat...

​What is machine learning? | Ask Google: The Machine Learning Edition

🔗 What is machine learning? | Ask Google: The Machine Learning Edition
You asked Google about machine learning. Our experts answered! In our Machine Learning Edition of Ask Google, we answer your most-searched questions about ma...

https://habr.com/ru/company/mailru/blog/438560/

Пишем XGBoost с нуля — часть 1: деревья решений

#machinelearning #neuralnets #deeplearning #машинноеобучение

Наш телеграмм канал - https://yangx.top/ai_machinelearning_big_data

TensorFlow.js Crash Course - Machine Learning For The Web - Getting Started

🎥 Learn TensorFlow.js - Deep Learning and Neural Networks with JavaScript
👁 1 раз ⏳ 4777 сек.
This full course introduces the concept of client-side artificial neural networks. We will learn how to deploy and run models along with full deep learning applications in the browser! To implement this cool capability, we’ll be using TensorFlow.js (TFJS), TensorFlow’s JavaScript library.

By the end of this video tutorial, you will have built and deployed a web application that runs a neural network in the browser to classify images! To get there, we'll learn about client-server deep learning architectures

Kaggle Human Protein: классификация паттернов белков — Дмитрий Буслов

🎥 Kaggle Human Protein: классификация паттернов белков — Дмитрий Буслов
👁 1 раз ⏳ 1188 сек.
Дмитрий Буслов рассказывает про задачу многоклассовой классификации изображений паттернов белков, которая решалась в рамках Kaggle Human Protein Atlas Image Classification, где его команде удалось войти в золото. Из видео вы сможете узнать про используемые архитектуры сетей и трюки обучения (в том числе неудачные), организацию второго уровня и лучшие решения контеста.

Слайды: https://gh.mltrainings.ru/presentations/Buslov_KaggleHumanProtein_2019.pdf

Узнать о текущих соревнованиях можно на сайте http://mlt

​New interesting paper in Super Resolution area

https://arxiv.org/abs/1812.04240

🔗 Unsupervised Degradation Learning for Single Image Super-Resolution
Deep Convolution Neural Networks (CNN) have achieved significant performance on single image super-resolution (SR) recently. However, existing CNN-based methods use artificially synthetic low-resolution (LR) and high-resolution (HR) image pairs to train networks, which cannot handle real-world cases since the degradation from HR to LR is much more complex than manually designed. To solve this problem, we propose a real-world LR images guided bi-cycle network for single image super-resolution, in which the bidirectional structural consistency is exploited to train both the degradation and SR reconstruction networks in an unsupervised way. Specifically, we propose a degradation network to model the real-world degradation process from HR to LR via generative adversarial networks, and these generated realistic LR images paired with real-world HR images are exploited for training the SR reconstruction network, forming the first cycle. Then in the second reverse cycle, consistency of real-world LR images are exploi

​Box and Whisker Plots of Mean Squared Error With Unscaled, Normalized and Standardized Input Variables for the Regression Problem
How to Improve Neural Network Stability and Modeling Performance With Data Scaling

🔗 How to Improve Neural Network Stability and Modeling Performance With Data Scaling
Deep learning neural networks learn how to map inputs to outputs from examples in a training dataset. The weights of the model are initialized to small random values and updated via an optimization algorithm in response to estimates of error on the training dataset. Given the use of small weights in the model and the …

Unsupervised learning for anomaly detection in stock options pricing
https://towardsdatascience.com/unsupervised-learning-for-anomaly-detection-in-stock-options-pricing-e599728958c7?source=collection_home---4------1---------------------

https://habr.com/ru/post/439350/

Искусственный интеллект как машина изобретений

#machinelearning #neuralnets #deeplearning #машинноеобучение

Наш телеграмм канал - https://yangx.top/ai_machinelearning_big_data

​How to Avoid Exploding Gradients in Neural Networks With Gradient Clipping

🔗 How to Avoid Exploding Gradients in Neural Networks With Gradient Clipping
Training a neural network can become unstable given the choice of error function, learning rate, or even the scale of the target variable. Large updates to weights during training can cause a numerical overflow or underflow often referred to as “exploding gradients.” The problem of exploding gradients is more common with recurrent neural networks, such …

Reinforcement Learning 8: Advanced Topics in Deep RL

🎥 Reinforcement Learning 8: Advanced Topics in Deep RL
👁 1 раз ⏳ 5314 сек.

🎥 MIT AI: Cruise Automation (Kyle Vogt)
👁 1 раз ⏳ 3324 сек.
Kyle Vogt is the President and CTO of Cruise Automation, leading an effort in trying to solve one of the biggest robotics challenges of our time: vehicle autonomy. He is the co-founder of 2 successful companies (Cruise and Twitch) that were each acquired for 1 billion dollars. This conversation is part of the Artificial Intelligence podcast and the MIT course 6.S094: Deep Learning for Self-Driving Cars. The conversation and lectures are free and open to everyone. Audio podcast version is available on https:

🎥 Introducing the modern data warehouse solution pattern with Azure SQL Data Warehouse
👁 1 раз ⏳ 1011 сек.
A deep look at the robust foundation for all enterprise analytics, spanning SQL queries to machine learning and AI. Charles Feddersen, Principal Program Mgr. for Azure SQL Data Warehouse, starts you off by looking at the core updates to Azure's SQL Data Warehouse, ranging from improvements to query performance to new capabilities with row-level security and workload importance. And finally, you'll learn how the Modern Data Warehouse Solution pattern, comprised of Data Factory, Data Lake Store, Databricks an

🎥 Reinforcement Learning 8: Advanced Topics in Deep RL
👁 1 раз ⏳ 5314 сек.

🎥 MIT 6.S191: Introduction to Deep Learning
👁 1 раз ⏳ 2728 сек.
MIT Introduction to Deep Learning 6.S191: Lecture 1
*New 2019 Edition*
Foundations of Deep Learning
Lecturer: Alexander Amini
January 2019


For all lectures, slides and lab materials: http://introtodeeplearning.com

https://vk.com/video-101965347_456262151?list=6af69056982daf477a
https://vk.com/video-101965347_456262152?list=7875cc5078d518ac38
https://vk.com/video-101965347_456262153?list=e8c65803973c847c94
https://vk.com/video-101965347_456262154?list=5823de3291c5268e09
https://vk.com/video-101965347_456262155?list=54bd82cf2e60e0ddb6
https://vk.com/video-101965347_456262156?list=8f163637899fb23d9b
https://vk.com/video-101965347_456262157?list=f369d7c67406bc1d6b
https://vk.com/video-101965347_456262158?list=c61976a5c8513f76f3

🎥 Лекция 1. Введение в нейронные сети.
👁 1829 раз ⏳ 2906 сек.


🎥 Лекция 2. Машинное обучение в эпоху больших данных.
👁 616 раз ⏳ 5775 сек.
Ссылка на слайды: https://drive.google.com/open?id=1NViPklG4RrOpb8XMMWETgHp6Qs4T0SuD

🎥 Лекция 3. Перцептрон и однослойные сети.
👁 370 раз ⏳ 2336 сек.
слайды лекции: https://vk.com/doc83865491_454180587?hash=37e49000cda3837459&dl=088a76281b090b67e5

🎥 Лекция 4. Backpropagation. Многослойные нейронные сети.
👁 361 раз ⏳ 5813 сек.


🎥 Лекция 5. Введение в свёрточные сети.
👁 296 раз ⏳ 5773 сек.


🎥 Лекция 6. Современные CNN архитектуры.
👁 207 раз ⏳ 4805 сек.
На этой лекции рассказывается об архитектурах свёрточных нейронных сетей, которые в своё время показали лучшие результаты на датасете ImageNet. Раз...

🎥 Лекция 7. Задачи детектирования и сегментации.
👁 244 раз ⏳ 2824 сек.
Слайды презентации: https://goo.gl/xHVsf3

🎥 Лекция 8. Введение в RNN
👁 207 раз ⏳ 1616 сек.
Лекция посвящена введению в рекуррентные нейронные сети.

​On Digital Irony, Sincerity and the Rhythm which Binds the Two*

🔗 On Digital Irony, Sincerity and the Rhythm which Binds the Two*
Amidst the rise of the far right and subsequent post-truth narratives in 2016, there have been multiple accounts which amplified what we…

​🔗 Academy on Air: Yпростить управление кампанией и повысить эффективность с помощью машинного обучения
учить больше: https://events.withgoogle.com/onlain-akadiemiia-rieklamy-2019/#content

Машинное обучение меняет подход компаний к бизнесу. Сегодня агентства могут уделять больше внимания долгосрочным планам развития, ведь теперь рутинные действия и оптимизация выполняются автоматически. В этой прямой трансляции мы покажем, как технологии машинного обучения помогают управлять кампаниями и достигать целей.

Swift for TensorFlow for Deep Learning in Google Colab: Train a CNN for MNIST Digit Classification

https://www.youtube.com/watch?v=_D-l7OGq83U

🎥 Swift for TensorFlow for Deep Learning in Google Colab: Train a CNN for MNIST Digit Classification
👁 1 раз ⏳ 2249 сек.
I hope you enjoyed this tutorial!

If you did, please make sure to leave a like, comment, and subscribe! It really does help out a lot!

Links:
Swift for TensorFlow on Github: https://github.com/tensorflow/swift
"tensorflow" branch on apple/swift on Github: https://github.com/apple/swift/tree/tensorflow
Swift for Tensorflow CIFAR example: https://github.com/tensorflow/swift-models/tree/stable/CIFAR
Setup Google Colab w/ Swift for TensorFlow: https://github.com/tensorflow/swift/blob/master/Usage.md
Predictin

​Как программист датасаентистам кернелы писал

Мало кто верит, что современный data science-стек может быть построен не на Python, но такие прецеденты есть :). Стек Одноклассников формировался долгие годы, в первую очередь программистами, перешедшими в data science, но всё ещё остались близкими к проду, поэтому в его основе лежат открытые технологии JVM-стека: Hadoop, Spark, Kafka, Cassandra и т.д. Это помогает нам сокращать время и затраты на ввод моделей в эксплуатацию, но иногда создаёт и сложности. Например, при подготовке базовых решений для участников SNA Hackathon 2019 пришлось сжать волю в кулак и погрузиться в мир динамической типизации. Подробности (и лёгкий троллинг) под катом :)

🔗 Как программист датасаентистам кернелы писал