Current Tricks in Deep Learning

Levente Szabados
Miklós Tóth

State-of-the-art Machine Learning / Deep Learning techniques

You don't have enough data in your project? Do you have an imbalanced dataset? Do you want to know how could you you get higher precision/recall for your model? Lower RMSE? The public datasets are really nice toys, but in real life, machine learning engineers have a tough time managing their data. At the same time, is also hard to follow the latest techniques and best practices for Deep Learning...In this hands-on workshop you can learn about the latest tricks and tips for Machine Learning / Deep Learning.

The presenters will show you:

- what to do if you don't have enough data

- how to deal with imbalanced datasets

- what are the latest activation functions for Neural Networks

- what are the state-of-the-art optimizers

- what are the newest loss functions?

The presenters are not only planning to discuss the theory behind these techniques but they will also demonstrate them with source codes, so the audience can watch these techniques in practice.

This workshop is recommended for developers with at least a basic knowledge with machine learning/Deep Learning. During the coding sessions tensorflow 2.0 (mainly the keras API) with Python will be used.

DB 401 - Hands on Deep Learning with Keras, TensorFlow, and Apache Spark™- official Databriks workshop

András Fülöp

Transparency, auditability, and stability of predictive models and results are typically key differentiators in effective machine learning applications. András will share tips and techniques learned through implementing interpretable machine learning solutions in industries like financial services, telecom, and health insurance. Using a set of publicly available and highly annotated examples, he teaches several holistic approaches to interpretable machine learning. The examples use the well-known University of California Irvine (UCI) credit card dataset and popular open source packages to train constrained, interpretable machine learning models and visualize, explain, and test more complex machine learning models in the context of an example credit-risk application. Along the way, András draws on his applied experience to highlight crucial success factors and common pitfalls not typically discussed in blog posts and open source software documentation, such as the importance of both local and global explanation and the approximate nature of nearly all machine learning explanation techniques.

Who is this presentation for?

Researchers, scientists, data analysts, predictive modelers, business users and other professionals, and anyone else who uses or consumes machine learning techniques

Prerequisite knowledge

A working knowledge of Python, widely used linear modeling approaches, and machine learning algorithms.

Materials or downloads needed in advance

A laptop with a recent version of the Firefox or Chrome browser installed. (This tutorial will use an Aquarium environment.) As a backup, tutorial materials are available on GitHub:

What you'll learn

The audience will learn several practical machine learning interpretability techniques and how to use them with Python. They will also learn the best way to use these techniques and common pitfalls to avoid when applying them.