November 13th, 2019

Useful Machine Learning Sessions from the H2O World New York

RSS icon RSS Category: H2O World, Machine Learning Interpretability, Makers

Conferences not only help us learn new skills but also enable us to build brand new relationships and networks along the way. H2O World is one such interactive community event featuring advancements in AI, machine learning, and explainable AI. It is a platform where people not only get to connect with the fantastic community but also learn to harness the full value of AI, machine learning, explainable AI, deep learning, and data science from industry-recognized speakers.

The recent edition of H2O World was held in New York and was attended by who’s who of the Data Science industry. With a stellar line up of speakers, experts, and Kaggle Grandmasters, the event was a huge success. There were several sessions and training conducted during the event. This article lists some of the important talks related to Driverless AI and H2O-3 from this H2O World conference. These talks/sessions are a great way to understand how Driverless AI and libraries like H2O can be utilized to solve some of the critical business problems with AI. Products and Solutions offers a range of AI and data science platforms from industry-leading open-source, H2O platform, to integrations for Apache Spark with Sparkling water, acceleration for NVIDIA GPU with H2O4GPU, and the award-winning H2O Driverless AI platform that delivers an expert data scientist in a box. provides leading AI technologies and products for every organization.

Follow along 💻

In case you want to follow along, you can take the 21 day trial of the Driverless AI called the Test Drive. This eliminates the need to download software, and you can learn how to build machine learning models using H2O Driverless AI on the AWS cloud.

There is also a great set of tutorials to help you get started easily with Driverless AI and H2O-3, through a step by step process. You can find them here:

Intro to Driverless AI

If you are thinking of getting your feet wet in Driverless AI, these videos are just what you need. David Whiting, Head of Global Training at a hands-on demonstration using the Credit Card dataset to predict which customer will default on the payment. This is a classic classification problem and a great use case for machine learning.

Time Series in H2O Driverless AI

Time series is a unique field in predictive modeling where standard feature engineering techniques and models are employed to get the most accurate results.

Dmitry Larko, a Senior Data Scientist at, gives a brief overview of Time Series data with Driverless AI. Dmitry showcases the capabilities of Driverless AI’s Time series recipe while also covering certain validation, feature engineering, feature selection, and modeling strategies.

Scalable Automatic Machine Learning with H2O

In this presentation, Erin LeDell, Chief Machine Learning Scientist,, provides a history and overview of the field of “Automatic Machine Learning” (AutoML), followed by a detailed look inside H2O’s open-source AutoML algorithm. The presentation also showcases a demo of H2O AutoML in R and Python, including a handful of code examples to get one started in automatic machine learning quickly.

Get Started with Driverless AI Recipes

Recipes are customizations and extensions to the Driverless AI platform. These can be custom machine learning models, transformers, or scorers (classification or regression), written in Python. Data scientists can bring their own recipes or leverage the open-source recipes available by the community and curated by data science experts.

Michelle Tanco, a Customer Solutions Engineer & Data Scientist for In this Hands-on session, she goes into the details of the recipes feature in Driverless AI and how can one create a recipe and integrate it with the Driverless AI platform.

TensorBoard integration and Image Recognition with Driverless AI

Yauhen Babakhina Data Scientist at H2O, shows us the capability of Driverless AI with image data. He demos an Image Recognition task using Driverless AI, where the task is to classify an image in one of the given six categories.

NLP with H2O Driverless AI

Sudalai Rajkumar, (aka SRK) is a Senior Data Scientist at In his talk, he throws light on the Natural Language capabilities of Driverless AI w.r.t text classification and regression problems. He discusses how Driverless AI can address a whole new set of problems in the text space like automatic document classification, sentiment analysis, emotion detection, and so on, using the textual data.

AutoDoc with H2O Driverless AI

Megan Kurka, Customer Data Scientist at H2O, talks about Driverless AI’s AutoDoc. AutoDoc enables a user to automatically document and explains the processes used by the Driverless AI platform. In a way, it frees up the user from summarizing their workflow while building machine learning models. The documentation includes details about the data used, the validation schema selected, model and feature tuning, and the final model created.

data.table for R and Python

Matt Dowle is the main author of the famous data.table package in R. data.table provides a high-performance version of R’s data.frame with syntax and feature enhancements for ease of use, convenience, and programming speed. In this presentation, Matt talks about the data.table package in R in detail. H2O has ported data.table to Python as datatable or Pydatatable, which like R’ data.table is also completely open-sourced.

The Case for Model Debugging

Patrick Hall is the Senior Director of Product at where he focuses mainly on model interpretability. This presentation outlines several standards and newer model debugging techniques and proposes several potential remediation methods for any discovered bugs. Discussed debugging techniques include adversarial examples, benchmark models, partial dependence and individual conditional expectation, random attacks, Shapley explanations of predictions and residuals, and models of residuals. Proposed remediation approaches include alternate models, editing of deployable model artifacts, missing value injection, prediction assertions, and regularization methods.

Additional Links

About the Author

Parul Pandey

Parul focuses on the intersection of, data science and community. She works as a Principal Data Scientist and is also a Kaggle Grandmaster in the Notebooks category.

Leave a Reply

H2O LLM DataStudio Part II: Convert Documents to QA Pairs for fine tuning of LLMs

Convert unstructured datasets to Question-answer pairs required for LLM fine-tuning and other downstream tasks with

September 22, 2023 - by Genevieve Richards, Tarique Hussain and Shivam Bansal
Building a Fraud Detection Model with H2O AI Cloud

In a previous article[1], we discussed how machine learning could be harnessed to mitigate fraud.

July 28, 2023 - by Asghar Ghorbani
A Look at the UniformRobust Method for Histogram Type

Tree-based algorithms, especially Gradient Boosting Machines (GBM's), are one of the most popular algorithms used.

July 25, 2023 - by Hannah Tillman and Megan Kurka
H2O LLM EvalGPT: A Comprehensive Tool for Evaluating Large Language Models

In an era where Large Language Models (LLMs) are rapidly gaining traction for diverse applications,

July 19, 2023 - by Srinivas Neppalli, Abhay Singhal and Michal Malohlava
Testing Large Language Model (LLM) Vulnerabilities Using Adversarial Attacks

Adversarial analysis seeks to explain a machine learning model by understanding locally what changes need

July 19, 2023 - by Kim Montgomery, Pramit Choudhary and Michal Malohlava
Reducing False Positives in Financial Transactions with AutoML

In an increasingly digital world, combating financial fraud is a high-stakes game. However, the systems

July 14, 2023 - by Asghar Ghorbani

Ready to see the platform in action?

Make data and AI deliver meaningful and significant value to your organization with our state-of-the-art AI platform.