Tips for Training Likelihood Models

This is a tutorial on common practices in training generative models that optimize likelihood directly, such as autoregressive models and normalizing flows.

http://bit.ly/2NQEmAq

Building Lyft’s Marketing Automation Platform

This post on Lyft's Engineering blog walks-through the machine learning system that enables Lyft's marketing at scale. It's fairly high-level but it's a good read and includes worthwhile details along the way.

https://lft.to/2NWlCiU

NumPy implementations of various ML models

Repository of mostly pure NumPy implementations of machine learning models. These are bare-bones implementations and aren't optimized to be efficient. They're optimized to be understanding how they work.

http://bit.ly/2lbReDe

​​Let's meet at AIUkraine 2019 in September!

AI Ukraine is an annual and most professional industry conference powered by AltexSoft.

The conference will include three stages:
- Data Science & Machine Learning
- BigData & Analytics
- Business & Startups

Special promo code for 7% discount for our subscribers - DSDigest-AI2019

Registration and more information: http://bit.ly/2O5DOXz

Top 25 pandas tricks

25 tricks that will help you to work faster and write better pandas code.

http://bit.ly/2O65qMn

Anomaly Detection for Dummies

This tutorial explores unsupervised anomaly detection for univariate and multivariate data. Covers a variety of detection strategies with python code snippets and screenshots.

Jupyter notebook: http://bit.ly/2O2PFp9

http://bit.ly/2O7unaa

TRFL

TRFL is a library built on top of TensorFlow that exposes several useful building blocks for implementing Reinforcement Learning agents, which was developed by the Research Engineering team at DeepMind.

http://bit.ly/2O8nfKu

Top 10 Statistics Mistakes Made by Data Scientists

Data scientists with no background in statistics are hardly a rarity today. Here are 10 mistakes that data scientists make using statistics, with examples and solutions included.

http://bit.ly/2Ockqbz

Building Data Pipelines With Kafka

To get value from this post, we recommend that you have Kafka installed. The article is for beginner engineers who are going to build their first data pipeline. If you are a seasoned pro, you might as well scan the article to freshen up your knowledge, too.

http://bit.ly/2OfsGaF

AI Behind LinkedIn Recruiter Search and Recommendation Systems

LinkedIn Recruiter is a LinkedIn’s key tool to help recruiters and hiring managers source suitable talent and identify “talent pools”. This blog post highlights a few unique challenges in information retrieval, system and recommender modeling, system design, architecture, and product deployment.

http://bit.ly/2Mb6VX2

Learning World Graphs to Accelerate Hierarchical Reinforcement Learning

An autonomous agent often counters various tasks within a single complex environment. Our two-stage framework proposes to first build a simple directed weighted graph abstraction over the world in an unsupervised task-agnostic manner and then to accelerate the hierarchical reinforcement learning of a diversity of downstream tasks.

Paper: http://bit.ly/2MfmWeq

http://bit.ly/2MfmXz0

MIT Deep Learning Basics: Introduction and Overview with TensorFlow

This blog post provides an overview of deep learning in 7 architectural paradigms with links to TensorFlow tutorials for each. It accompanies the following lecture on Deep Learning Basics as part of MIT course 6.S094.

http://bit.ly/2Mnf1vs

Personalized Recommendations for Experiences Using Deep Learning

In this blog post, you will learn how TripAdvisor’s newly-developed ‘Recommended For You’ (RFY) model generates personalized recommendations on their website using users’ browsing history and deep learning.

http://bit.ly/2MgiDPU

Introducing Dagster

Dagster is an open-source Python library for building systems like ETL processes and ML pipelines.

GitHub: http://bit.ly/2MkyYTP

http://bit.ly/2Mkw4yh

Python Machine Learning Tutorial: Predicting Airbnb Prices

This tutorial will introduce you to the fundamental concepts of machine learning. As you follow along, you’ll build your very first model from scratch to make predictions while developing a solid understanding of how exactly how your model works.

http://bit.ly/2Mn39JZ

Using DVC to create an efficient version control system for data projects

In this article, you will learn what DVC is and how to use it to track the project’s data. DVC simplifies Data projects by using stages and pipelines, which increases productivity and improves collaboration.

http://bit.ly/2MvgLmN

ICML 2019 Videos

Video and slides from the International Conference on Machine Learning

http://bit.ly/2ys9uvG

How to Develop an Information Maximizing GAN (InfoGAN) in Keras

In this tutorial, you will learn how to develop and train an InfoGAN model from scratch and use the control variables to control which digit is generated by the model.

http://bit.ly/2MxWpJB

#ML #AI #DataScience

Neural Architecture Search at CVPR 2019

In this article, you will learn about neural architecture search (NAS) and how it was presented at CVPR 2019 in Long Beach.

http://bit.ly/333nSbH

PyTorch Transformers for state-of-the-art NLP

If you're doing anything with NLP, this is a great new open-source tool to know about. This library from Hugging Face contains 27 pre-trained models to conduct state-of-the-art NLP/NLU tasks, including BERT, GPT-2, XLNet, etc. It's a unified API too, which makes it easy to use and experiment with the latest techniques.

http://bit.ly/337Mgch

Cyclical Learning Rates with Keras and Deep Learning

In this tutorial, you will learn how to use Cyclical Learning Rates (CLR) and Keras to train your own neural networks. Using Cyclical Learning Rates you can dramatically reduce the number of experiments required to tune and find an optimal learning rate for your model.

http://bit.ly/2MEAAbc