​Распознавание эмоций с помощью сверточной нейронной сети
Блог компании Voximplant,

Распознавание эмоций всегда было захватывающей задачей для ученых. В последнее время я работаю над экспериментальным SER-проектом (Speech Emotion Recognition), чтобы понять потенциал этой технологии – для этого я отобрал наиболее популярные репозитории на Github и сделал их основой моего проекта.

Прежде чем мы начнем разбираться в проекте, неплохо будет вспомнить, какие узкие места есть у SER.
Программирование,
Машинное обучение,
Искусственный интеллект,
https://habr.com/ru/company/Voximplant/blog/461435/

🔗 Распознавание эмоций с помощью сверточной нейронной сети
Распознавание эмоций всегда было захватывающей задачей для ученых. В последнее время я работаю над экспериментальным SER-проектом (Speech Emotion Recognition),...

​Differentiable Programming — Inverse Graphics AutoEncoder
Let’s consider handwritten character recognition using MNIST (EMNIST).

https://towardsdatascience.com/differentiable-programming-inverse-graphics-autoencoder-e1b0fabe67bf?source=collection_home---4------3-----------------------

🔗 Differentiable Programming — Inverse Graphics AutoEncoder
Let’s consider handwritten character recognition using MNIST (EMNIST).

​How to Communicate Clearly About Machine Learning.
Your choice of words matters.
https://towardsdatascience.com/how-to-communicate-clearly-about-machine-learning-8731e4d1cd4c?source=collection_home---4------1-----------------------

🔗 How to Communicate Clearly About Machine Learning.
Your choice of words matters.

​Predicting vs. Explaining
And Why Data Science Needs More “Half-Bayesians”
https://towardsdatascience.com/predicting-vs-explaining-69b516f90796?source=collection_home---4------0-----------------------

🔗 Predicting vs. Explaining
And Why Data Science Needs More “Half-Bayesians”

PyTorch Tutorial - Deep Learning Using PyTorch - Learn PyTorch from Basics to Advanced

Наш телеграм канал - tglink.me/ai_machinelearning_big_data

https://www.youtube.com/watch?v=LeCZyHE5vlk

🎥 PyTorch Tutorial - Deep Learning Using PyTorch - Learn PyTorch from Basics to Advanced
👁 1 раз ⏳ 5602 сек.
Learn PyTorch from the very basics to advanced models like Generative Adverserial Networks and Image Captioning
A Complete Machine Learning Project Walk-Through in Python
☞ https://morioh.com/p/b56ae6b04ffc

Deep Learning With TensorFlow 2.0
☞ https://morioh.com/p/d669c3deea75

Introduction to PyTorch and Machine Learning
☞ https://morioh.com/p/296b2e812203

Machine Learning A-Z™: Hands-On Python & R In Data Science
☞ http://learnstartup.net/p/SJw1YoTMg

Deep Learning A-Z™: Hands-On Artificial Neural Networ

​Eidetic 3D LSTM: A Model for Video Prediction and Beyond

https://openreview.net/forum?id=B1lKS2AqtX

Git : https://github.com/metrofun/E3D-LSTM

🔗 Eidetic 3D LSTM: A Model for Video Prediction and Beyond
Spatiotemporal predictive learning, though long considered to be a promising self-supervised feature learning method, seldom shows its effectiveness beyond future video prediction. The reason is...

​How to screw up a Computer Vision project
https://medium.com/@pjgrizel/how-to-screw-up-a-computer-vision-project-166dfcc44a5f

🔗 How to screw up a Computer Vision project
Statistics (are they accurate at all?) tell that 85% of AI projects do not deliver. That leaves only 15% real successes, and they are good…

Deep convolutional neural networks for uncertainty propagation in random fields
The development of a reliable and robust surrogate model is often constrained by the dimensionality of the problem. For a system with high-dimensional inputs/outputs (I/O), conventional approaches usually use a low-dimensional manifold to describe the high-dimensional system, where the I/O data is first reduced to more manageable dimensions and then the condensed representation is used for surrogate modeling. In this study, we present a new solution scheme for this type of problems based on a deep learning approach.
https://arxiv.org/abs/1907.11198

Машинное обучение (2019) (Часть 1)

Наш телеграм канал - tglink.me/ai_machinelearning_big_data
01 - Задачи и модели машинного обучения
02 - Проблемы машинного обучения
03 - Представление данных для машинного обучения. Признаки объектов
04 - Основные понятия математической статистики. Часть 1
05 - Основные понятия математической статистики. Часть 2
06 - Постановка проблемы и простейшие способы ее решения
07 - Восстановление данных с помощью метрики
08 - Замечание об использовании метрики
09 - Использование коэффициента корреляции для восстановления данных
10 - Применение метрик и КК в рекомендательных системах

🎥 01 - Машинное обучение. Задачи и модели машинного обучения
👁 1 раз ⏳ 484 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

🎥 02 - Машинное обучение. Проблемы машинного обучения
👁 1 раз ⏳ 333 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

🎥 03 - Машинное обучение. Представление данных для машинного обучения. Признаки объектов
👁 1 раз ⏳ 389 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

🎥 04 - Машинное обучение. Основные понятия математической статистики. Часть 1
👁 1 раз ⏳ 427 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

🎥 05 - Машинное обучение. Основные понятия математической статистики. Часть 2
👁 1 раз ⏳ 405 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

🎥 06 - Машинное обучение. Постановка проблемы и простейшие способы ее решения
👁 1 раз ⏳ 336 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

🎥 07 - Машинное обучение. Восстановление данных с помощью метрики
👁 1 раз ⏳ 768 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

🎥 08 - Машинное обучение. Замечание об использовании метрики
👁 1 раз ⏳ 397 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

🎥 09 - Машинное обучение. Использование коэффициента корреляции для восстановления данных
👁 1 раз ⏳ 269 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

🎥 10 - Машинное обучение. Применение метрик и КК в рекомендательных системах
👁 1 раз ⏳ 369 сек.
Лектор: Артём Шевляков

https://stepik.org/8057

Лекции по анализу данных от Технострим
tglink.me/pythonl - наш телеграм канал
1 - Анализ данных. Введение в python
2 - Анализ данных. Advanced Python
3 - Анализ данных. Библиотеки Python
4 - Анализ данных. Визуализация, анализ датасета EDA
5 - Анализ данных. R и библиотеки
6 - Анализ данных. Введение в статистику
7 - Анализ данных. Статистическое оценивание
8 - Анализ данных. Параметрические статистические тесты
9 - Анализ данных. Непараметрические тесты
10 - Анализ данных. Множественная проверка гипотез
https://vk.com/video-16108331_456263704?list=5e96c39aeb208bec64
https://vk.com/video-16108331_456263705?list=1d38209df3d26fba3e
https://vk.com/video-16108331_456263706?list=57fbed9df30695f358
https://vk.com/video-16108331_456263707?list=0f1da65f04176ec8f3
https://vk.com/video-16108331_456263708?list=246551a44e7050d2b1
https://vk.com/video-16108331_456263709?list=69a241025a7619b6d7
https://vk.com/video-16108331_456263710?list=57cbebdbfdae471afc
https://vk.com/video-16108331_456263711?list=d5f7331db4927b6405
https://vk.com/video-16108331_456263712?list=7ba545f2462cb93c47
https://vk.com/video-16108331_456263713?list=3d967e2ec1bdf288e6

🎥 1 - Анализ данных. Введение в python
👁 1 раз ⏳ 7114 сек.
Техносфера Mail.ru Group, МГУ им. М.В. Ломоносова
Курс "Введение в анализ данных"
Источник: https://www.youtube.com/channel/UCmqEpAsQMcsYaeef4qgECvQ

🎥 2 - Анализ данных. Advanced Python
👁 1 раз ⏳ 4828 сек.
Техносфера Mail.ru Group, МГУ им. М.В. Ломоносова
Курс "Введение в анализ данных"
Источник: https://www.youtube.com/channel/UCmqEpAsQMcsYaeef4qgECvQ

🎥 3 - Анализ данных. Библиотеки Python
👁 1 раз ⏳ 4752 сек.
Техносфера Mail.ru Group, МГУ им. М.В. Ломоносова
Курс "Введение в анализ данных"
Источник: https://www.youtube.com/channel/UCmqEpAsQMcsYaeef4qgECvQ

🎥 4 - Анализ данных. Визуализация, анализ датасета EDA
👁 1 раз ⏳ 8094 сек.
Техносфера Mail.ru Group, МГУ им. М.В. Ломоносова
Курс "Введение в анализ данных"
Источник: https://www.youtube.com/channel/UCmqEpAsQMcsYaeef4qgECvQ

🎥 5 - Анализ данных. R и библиотеки
👁 1 раз ⏳ 5092 сек.
Техносфера Mail.ru Group, МГУ им. М.В. Ломоносова
Курс "Введение в анализ данных"
Источник: https://www.youtube.com/channel/UCmqEpAsQMcsYaeef4qgECvQ

🎥 6 - Анализ данных. Введение в статистику
👁 1 раз ⏳ 7046 сек.
Техносфера Mail.ru Group, МГУ им. М.В. Ломоносова
Курс "Введение в анализ данных"
Источник: https://www.youtube.com/channel/UCmqEpAsQMcsYaeef4qgECvQ

🎥 7 - Анализ данных. Статистическое оценивание
👁 1 раз ⏳ 5088 сек.
Техносфера Mail.ru Group, МГУ им. М.В. Ломоносова
Курс "Введение в анализ данных"
Источник: https://www.youtube.com/channel/UCmqEpAsQMcsYaeef4qgECvQ

🎥 8 - Анализ данных. Параметрические статистические тесты
👁 1 раз ⏳ 5274 сек.
Техносфера Mail.ru Group, МГУ им. М.В. Ломоносова
Курс "Введение в анализ данных"
Источник: https://www.youtube.com/channel/UCmqEpAsQMcsYaeef4qgECvQ

🎥 9 - Анализ данных. Непараметрические тесты
👁 1 раз ⏳ 6229 сек.


🎥 10 - Анализ данных. Множественная проверка гипотез
👁 1 раз ⏳ 4536 сек.

​AI Creates Near Perfect Images Of People, Dogs and More

🔗 AI Creates Near Perfect Images Of People, Dogs and More
❤️ Check out Weights & Biases here and sign up for a free demo: https://www.wandb.com/papers 📝 The paper "Generating Diverse High-Fidelity Images with VQ-VAE-2" and its supplementary materials are available here: https://arxiv.org/abs/1906.00446 https://drive.google.com/file/d/1H2nr_Cu7OK18tRemsWn_6o5DGMNYentM/view Our latent-space material synthesis paper is available here: https://users.cg.tuwien.ac.at/zsolnai/gfx/gaussian-material-synthesis/ 🙏 We would like to thank our generous Patreon supporters who

​Awesome Fraud Detection Research Papers

https://github.com/benedekrozemberczki/awesome-fraud-detection-papers

🔗 benedekrozemberczki/awesome-fraud-detection-papers
A curated list of data mining papers about fraud detection. - benedekrozemberczki/awesome-fraud-detection-papers

State of Data Science & Machine Learning - Peter Wang
https://www.youtube.com/watch?v=S3FpT4xMyn4

🎥 State of Data Science & Machine Learning - Peter Wang
👁 1 раз ⏳ 2292 сек.
As machine learning and AI become adopted at an increasing rate, businesses and practitioners face new types of challenges. At the heart of many of these lies an uncomfortable truth: that data science is not merely a new kind of technical specialty, but rather that it represents an opportunity for deep business transformation. In this talk, Peter speaks to this concept that Data Science isn’t just a “job”, it’s actually a democratization of empiricism. Furthermore, the idea of “democratization” is intertwin

​Simple and multiple linear regression with Python
Linear regression is a linear approach to model the relationship between a dependent variable
https://towardsdatascience.com/simple-and-multiple-linear-regression-with-python-c9ab422ec29c?source=collection_home---4------1-----------------------

🔗 Simple and multiple linear regression with Python
Linear regression is a linear approach to model the relationship between a dependent variable (target variable) and one (simple…

​How to Train Your Quadcopter
The beginner’s guide to teaching a quadcopter to fly (with code!)
https://towardsdatascience.com/how-to-train-your-quadcopter-adventures-in-machine-learning-algorithms-e6ee5033fd61?source=collection_home---4------0-----------------------

🔗 How to Train Your Quadcopter
The beginner’s guide to teaching a quadcopter to fly (with code!)

​The theory you need to know before you start an NLP project
An overview of the most common natural language processing and machine learning techniques needed to start tackling any project involving
https://towardsdatascience.com/the-theory-you-need-to-know-before-you-start-an-nlp-project-1890f5bbb793?source=collection_home---4------0-----------------------

🔗 The theory you need to know before you start an NLP project
An overview of the most common natural language processing and machine learning techniques needed to start tackling any project involving…

OpenCV Python Tutorial For Beginners 28 - Hough Line Transform Theory
https://www.youtube.com/watch?v=7m-RVJ6ABsY

🎥 OpenCV Python Tutorial For Beginners 28 - Hough Line Transform Theory
👁 1 раз ⏳ 759 сек.
In this video on OpenCV Python Tutorial For Beginners, we are going to understand the concept of the Hough Transform and Hough Line Transform Theory.

OpenCV is an image processing library created by Intel and later supported by Willow Garage and now maintained by Itseez. opencv is available on Mac, Windows, Linux. Works in C, C++, and Python.
it is Open Source and free. opencv is easy to use and install.

Starting with an overview of what the course will be covering, we move on to discussing morphological

Linear Regression - Introduction to Machine Learning using Python and Scikit Learn Chapter 6 1
https://www.youtube.com/watch?v=3Q6gzUPecLE

🎥 Linear Regression - Introduction to Machine Learning using Python and Scikit Learn Chapter 6 1
👁 1 раз ⏳ 663 сек.
Welcome to the video series on Introduction to Machine Learning with Scikit-Learn.

This video contains Chapter - 6.1. In this chapter, I've explained our first Machine Learning algorithm called Linear Regression using just five data points for easy understanding

This video describes what is Linear Regression and how we can use the same using Scikit-learn. In context of this algorithm, I've also explained the unified machine learning algorithm and how generic interface can be used for almost all ML algori

​Visualizing Keras Models
I love how simple and clear Keras makes it to build neural networks. With wandb, you can visualize your network’s performance
https://towardsdatascience.com/visualizing-keras-models-49d591931209?source=topic_page---------42------------------1

🔗 Lightweight Visualization of Keras Models
I love how simple and clear Keras makes it to build neural networks. With wandb, you can visualize your network’s performance and…

​Deep convolutional neural networks for uncertainty propagation in random fields

Authors: Xihaier Luo, Ahsan Kareem

Abstract: …manageable dimensions and then the condensed representation is used for surrogate modeling. In this study, we present a new solution scheme for this type of problems based on a deep learning approach. The proposed surrogate is based on a particular network architecture, i.e. the convolutional neural networks.
https://arxiv.org/abs/1907.11198

🔗 Deep convolutional neural networks for uncertainty propagation in random fields
The development of a reliable and robust surrogate model is often constrained by the dimensionality of the problem. For a system with high-dimensional inputs/outputs (I/O), conventional approaches usually use a low-dimensional manifold to describe the high-dimensional system, where the I/O data is first reduced to more manageable dimensions and then the condensed representation is used for surrogate modeling. In this study, we present a new solution scheme for this type of problems based on a deep learning approach. The proposed surrogate is based on a particular network architecture, i.e. the convolutional neural networks. The surrogate architecture is designed in a hierarchical style containing three different levels of model structures, advancing the efficiency and effectiveness of the model in the aspect of training and deploying. To assess the model performance, we carry out uncertainty quantification in a continuum mechanics benchmark problem. Numerical results suggest the proposed model is capable of d