H2O.ai's Driverless AI

•

6 likes•2,727 views

The document discusses H2O.ai's Driverless AI product, which aims to automate and simplify the machine learning process. It provides an overview of H2O.ai as a company, their goals of operationalizing data science. Driverless AI uses techniques like automated feature engineering, model tuning and selection, and model ensembling to build accurate models fast. It also allows for interpreting and explaining machine learning models through features like model inspection and reason codes. A demo of Driverless AI predicting credit card default risk is shown to illustrate the system.

H2O.ai Driverless AI
Using AI to do Fast, Accurate and Interpretable AI
Jo-fai (Joe) Chow
Data Scientist
joe@h2o.ai
@matlabulous
Big Data LDN
15th November, 2017

Agenda
• Introduction
• Company / People
• Using AI to do AI
• Evolution of H2O Platform
• Motivation
• Driverless AI Key Features
• Credit Risk Demo
• Other News
2

Company Overview
Founded 2011 Venture-backed, debuted in 2012
Products • H2O Open Source In-Memory AI Prediction Engine
• Sparkling Water (H2O + Spark)
• Enterprise Steam
• Driverless AI
Mission Operationalize Data Science, and provide a platform for users to build beautiful data products
Team 75+ employees
• Distributed Systems Engineers doing Machine Learning
• World-class visualization designers
Headquarters Mountain View, CA
3

5
http://www.pwc.com/gx/en/about/stories-from-across-the-world/harnessing-the-power-of-ai-to-transform-the-detection-of-fraud-and-error.html
Award-Winning Application (PwC + H2O.ai)

Using AI to do AI
H2O.ai Driverless AI
7

H2O AI Platform Timeline
8
Users
Advanced Data Scientists
Developers/Engineers
Dev Ops
App Developers
H2O Core
Data.table
Sparkling Water
Roadmap
Steam
Visual
Interpretation
Deep Water
Analysts
H2O
Driverless AI
2012 2014 2016 2017 2018
H2O GPU Edition
GPU
2019
Auto ML

“Confidential and property of H2O.ai. All rights reserved”
Shortage of Data Scientists
“The United States alone faces a shortage of 140,000 to 190,000
people with analytical expertise and 1.5 million managers and
analysts”
–McKinsey Prediction for 2018

The Product: Driverless AI
• Kaggle Grandmasters in a Box
• A solution to the shortage of data scientists
• Fast
• H2O Algorithms optimised for GPUs
• Accurate
• Auto Feature Engineering
• Auto Model Tuning / Selection / Ensemble
• Interpretable
• Machine Learning Interpretation
• Production Ready
• Export pipeline as a standalone Python
package
11

Fast: H2O Algorithms on GPUs
12
3400+ models
were trained on
GPUs by the time
the first model
completed training
on CPUs
https://www.youtube.com/watch?v=LrC3mBNG7WU

Accurate: Auto Feature Engineering
13
Feature Transformation
Pipeline
Model
Training
Feature Engineering

Accurate: Auto Feature Engineering
14
• Cross Validation
Categorical Encoding
• Frequency Encoding
• Cross Validation Target
Encoding
• Truncated SVD and More
Feature Transformations
Generated
Features
Original Features

Accurate: Auto Model Tuning/Selection/Ensemble
15

Interpretable: Explainable Machine Learning
16
https://www.oreilly.com/ideas/ideas-on-interpreting-machine-learning

19
Training
Data
Target
Simple
Settings
Select Metric
That’s it!
Optional
Test Set

20
Auto Feature
Engineering
Trying out Different
Sets of Features
Using both CPUs & GPUs

21
Options for Interpretation,
Scoring, Logs, Pipeline

23
Zoom into
individual records
Explain how inputs
lead to outcome
with a simple
decision tree

External Review
27
https://www.infoworld.com/article/3236048/machine-learning/review-h2oai-automates-machine-learning.html

www.h2o.ai/driverless-ai-download/
28
30-Day Free TrialDocker Image

Using AI to do AI
H2O.ai Driverless AI
29

Thank you everyone for your support!
32
4000+ Members
450+ RSVPs!!!

London AI & Deep Learning Meetup
33
15th November, 2017
Keynote Theatre, Olympia - London
Time Session Title Speaker
6:00 – 6:30 pm
Doors Open
Pizzas, Drinks & Networking
/
6:30 – 6:45 pm Introduction Joe
6:45 – 7:30 pm
From Kagglenoobs to Kaggle
Master
Yifan
7:30 – 7:45 pm Break /
7:45 – 8:30 pm
Scaling Machine Learning
with H2O Sparkling Water
Kuba
8:30 – 9:00 pm Q&A and Networking /
Sponsored by

• Organisers
34
Thank you!
• Code, Slides & Documents
• bit.ly/h2o_meetups
• docs.h2o.ai
• Contact
• joe@h2o.ai
• @matlabulous
• github.com/woobe
• Please search/ask questions on
Stack Overflow
• Use the tag `h2o` (not H2 zero)

What's hot

Python: Basic Inheritance

Damian T. Gordon

The document discusses basic inheritance in Python. It explains that all classes inherit from the base Object class. Inheritance allows creating subclasses that inherit attributes and methods from a parent superclass. This allows code reuse and adding specialized behavior. An example creates a Contact class to store names and emails, with a Supplier subclass that adds an "order" method. A SupplierCheck subclass then overrides the order method to check the customer balance before processing orders.

RPA and Artificial Intelligence Brochure EQ - U2 b

Zoe Gammie

This document provides an agenda for the RPA and Artificial Intelligence Summit taking place from November 30th to December 2nd in London. The summit will bring together over 200 attendees from over 100 companies to discuss best practices for implementing robotic process automation (RPA) and artificial intelligence (AI) technologies. The agenda includes keynote sessions on RPA and AI use cases, panels with practitioners from various industries, and workshops on developing successful automation programs and pilots. Attendees will learn how to identify processes for automation, establish Centers of Excellence, address common challenges, and develop action plans to drive automation initiatives at their organizations. The event aims to help attendees augment human workforces with intelligent technologies.

Integration between Filebeat and logstash

DaeMyung Kang

Filebeat sends log files to Logstash. There are several cases described for integrating Filebeat and Logstash: 1) A simple configuration where one log file is sent from Filebeat to Logstash and output to one file. 2) Another simple configuration where multiple log files are sent from Filebeat to Logstash using a wildcard, and output to one file. 3) An advanced configuration where multiple log files are sent from Filebeat to Logstash, and Logstash outputs each file to a separate file based on the original file name using filtering. 4) A more advanced configuration where log files are sent from Filebeat to Logstash, Logstash parses the timestamp and uses it as the output

How Deloitte Uses AI to Simplify Reporting and Increase Value

Amazon Web Services

In this webinar, you’ll learn how Deloitte, a multinational consultancy, solves client pain points around employee efficiency, scale, regulatory compliance, and customer engagement using Quill, a NLG solution created by Narrative Science and deployed on AWS. You’ll learn how you can easily get started with artificial intelligence on AWS to derive deeper insights, improve operational efficiency, enhance customer experiences, and meet compliance requirements. Join our webinar to learn: - How NLG can solve problems around internal reporting, operational efficiency, and regulatory compliance - How Deloitte delivered transformative solutions both internally and to clients on AWS while saving over $600K - How to get started with NLG in your organization

The Future is in Responsible Generative AI

Saeed Al Dhaheri

Introduction to Artificial Intelligence on AWS

Amazon Web Services

This document provides an overview of artificial intelligence and machine learning services available on Amazon Web Services. It discusses how AI can power applications through services like Amazon Polly for text-to-speech, Amazon Lex for conversational interfaces, Amazon Rekognition for image and video analysis, and Amazon's deep learning platforms. It also provides examples of how customers in various industries like media, public safety, and marketing are using these AWS AI services.

Suresh Poopandi_Generative AI On AWS-MidWestCommunityDay-Final.pdf

AWS Chicago

The document discusses generative AI and tools for building with it on AWS. It provides an introduction to generative AI, describes common use cases like text generation and image generation, and reviews tools available on AWS for generative AI like Amazon Bedrock, Amazon EC2 Trn1n and Amazon EC2 Inf2, Amazon CodeWhisperer, and Amazon SageMaker Jumpstart. It also discusses security, customization, and cost benefits of using AWS for generative AI projects.

Kai Wang - AI for Innovation1.1r.pdf

SOLTUIONSpeople, THINKubators, THINKathons

This document discusses using AI for innovation and idea generation. It summarizes several studies that compare human and AI performance on divergent thinking and business idea generation tasks. The studies found that while AI can generate useful ideas, humans tend to be more innovative. The document also discusses how AI can be used as innovation tools to explore problems and generate solutions, but that overreliance on AI risks deskilling humans. It emphasizes that human expertise is still needed to effectively use AI for creativity.

Understanding GenAI/LLM and What is Google Offering - Felix Goh

NUS-ISS

With the recent buzz on Generative AI & Large Language Models, the question is to what extent can these technologies be applied at work or when you're studying and how easy is it to manage/develop your own models? Hear from our guest speaker from Google as he shares some insights into how industries are evolving with these trends and what are some of Google's offerings from Duet AI in Google Workspace to the GenAI App Builder on Google Cloud.

BDA301 An Introduction to Amazon Rekognition

Amazon Web Services

Evolution of AI in workplace.pptx

AshishUpadhyay957206

The document discusses the evolution of artificial intelligence (AI) and its impact on workplaces. It outlines three phases of AI development: artificial narrow intelligence, artificial general intelligence, and artificial super intelligence. Currently, AI is in the early stages of artificial general intelligence where machine and human intelligence are becoming more equal. The document then provides examples of useful AI tools for content writing, designing, and content creation/work that are augmenting human capabilities in the workplace. It concludes by anticipating greater integration of AI assistants like GitHub Copilot and Microsoft 365 Copilot in the future.

Amazon QuickSight First Call Deck

Amazon Web Services

Generative AI

lutzsuarnaba1

The document discusses how generative AI can be used to scale content operations by reducing the time it takes to generate content. It explains that generative AI learns from natural language models and can generate new text or ideas based on prompts provided by users. While generative AI has benefits like speeding up content creation and ideation, it also has limitations such as not being able to conduct original research or ensure quality. The document provides examples of how generative AI can be used for tasks like generating ideas, simplifying complex text, creating visuals, and more. It also discusses challenges like bias in AI models and the low risk of plagiarism.

The Future of AI is Generative not Discriminative 5/26/2021

Steve Omohundro

The deep learning AI revolution has been sweeping the world for a decade now. Deep neural nets are routinely used for tasks like translation, fraud detection, and image classification. PwC estimates that they will create $15.7 trillion/year of value by 2030. But most current networks are "discriminative" in that they directly map inputs to predictions. This type of model requires lots of training examples, doesn't generalize well outside of its training set, creates inscrutable representations, is subject to adversarial examples, and makes knowledge transfer difficult. People, in contrast, can learn from just a few examples, generalize far beyond their experience, and can easily transfer and reuse knowledge. In recent years, new kinds of "generative" AI models have begun to exhibit these desirable human characteristics. They represent the causal generative processes by which the data is created and can be compositional, compact, and directly interpretable. Generative AI systems that assist people can model their needs and desires and interact with empathy. Their adaptability to changing circumstances will likely be required by rapidly changing AI-driven business and social systems. Generative AI will be the engine of future AI innovation.

Building Data Lakes and Analytics on AWS; Patterns and Best Practices - BDA30...

Amazon Web Services

In this session, we show you how to understand what data you have, how to drive insights, and how to make predictions using purpose-built AWS services. Learn about the common pitfalls of building data lakes, and discover how to successfully drive analytics and insights from your data. Also learn how services such as Amazon S3, AWS Glue, Amazon Redshift, Amazon Athena, Amazon EMR, Amazon Kinesis, and Amazon ML services work together to build a successful data lake for various roles, including data scientists and business users.

Fraud Detection and Prevention on AWS using Machine Learning

Amazon Web Services

Fraud Detection and Prevention on AWS Speaker: Andrew Kane, Solutions Architect, AWS AI and real-time analytics are increasingly being used to quickly identify financial fraud. In this session, we will also give an overview of how to build and train a model to identify fraudulent card transactions from bona fide ones. We will also describe an architecture that allows you to feed live transaction data into a Machine Learning model to provide a real-time authentic / fraudulent classification of that transaction.

Python Scipy Numpy

Girish Khanzode

This document provides an overview of the Python programming language. It discusses Python's history and evolution, its key features like being object-oriented, open source, portable, having dynamic typing and built-in types/tools. It also covers Python's use for numeric processing with libraries like NumPy and SciPy. The document explains how to use Python interactively from the command line and as scripts. It describes Python's basic data types like integers, floats, strings, lists, tuples and dictionaries as well as common operations on these types.

Myths and Mathemagical Superpowers of Data Scientists

David Pittman

1) The document discusses 10 myths about data scientists and provides realities to counter each myth. 2) Some myths include claims that data scientists are mythical beings, elitist academics, or a fading trend. However, the realities note data science requires hands-on work with data and has experienced steady growth. 3) Other myths suggest data scientists are just statisticians or BI specialists, but the realities indicate data scientists come from varied backgrounds and tackle business problems through experimentation and analysis.

Unlocking the Power of Generative AI An Executive's Guide.pdf

PremNaraindas1

Generative AI is here, and it can revolutionize your business. With its powerful capabilities, this technology can help companies create more efficient processes, unlock new insights from data, and drive innovation. But how do you make the most of these opportunities? This guide will provide you with the information and resources needed to understand the ins and outs of Generative AI, so you can make informed decisions and capitalize on the potential. It covers important topics such as strategies for leveraging large language models, optimizing MLOps processes, and best practices for building with Generative AI.

Global Governance of Generative AI: The Right Way Forward

Lilian Edwards

AI regulation has been a hot topic since the rise of machine learning (ML) in the “big data” era, but generative AI or “foundation models” tools like ChatGPT, DALL-E 2(now 3) and CoPilot, ike ML before them, may create serious societal risks, including embedding and outputting bias; generating fake news, illegal or harmful content and inadvertent “hallucinations”; infringing existing laws relating eg to copyright and privacy; as well as environmental, competition and workplace concerns. Many nations are now considering regulation to address these worries, and can draw on a number of basic and hybrid models of governance. This paper canvasses models of mandatory comprehensive legislation (where the EU AI Act hopes to place itself as a gold standard model); vertical mandatory legislation (where China has quietly taken a lead); adapting existing law (see the many copyright lawsuits underway); and voluntary “soft law” such as codes of ethics, “blueprints”, or industry guidelines. Both the domestic and international regulatory scenes for AI are also increasingly politicised as the rise of "AI safety" hype shows. Against this backdrop what choices should smaller countries such as the UK and Australia make? will international harmonisation lead to a race to the top as with the GDPR, or the bottom - rule by tech for tech?

What's hot (20)

Python: Basic Inheritance

RPA and Artificial Intelligence Brochure EQ - U2 b

Integration between Filebeat and logstash

How Deloitte Uses AI to Simplify Reporting and Increase Value

The Future is in Responsible Generative AI

Introduction to Artificial Intelligence on AWS

Suresh Poopandi_Generative AI On AWS-MidWestCommunityDay-Final.pdf

Kai Wang - AI for Innovation1.1r.pdf

Understanding GenAI/LLM and What is Google Offering - Felix Goh

BDA301 An Introduction to Amazon Rekognition

Evolution of AI in workplace.pptx

Amazon QuickSight First Call Deck

Generative AI

The Future of AI is Generative not Discriminative 5/26/2021

Building Data Lakes and Analytics on AWS; Patterns and Best Practices - BDA30...

Fraud Detection and Prevention on AWS using Machine Learning

Python Scipy Numpy

Myths and Mathemagical Superpowers of Data Scientists

Unlocking the Power of Generative AI An Executive's Guide.pdf

Global Governance of Generative AI: The Right Way Forward

Viewers also liked

GPU Accelerated Machine Learning

H2O.ai's Driverless AI

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (7)

Similar to H2O.ai's Driverless AI

Similar to H2O.ai's Driverless AI (20)

More from Sri Ambati

More from Sri Ambati (20)

Recently uploaded

Recently uploaded (20)

H2O.ai's Driverless AI