With so much artificial intelligence (AI) in the headlines, it is no surprise that businesses are scrambling to exploit this exciting and transformative technology. Clearly, those who are the first to deliver business-relevant AI will gain significant advantage.
However, there is a problem! Our perception of AI success in society is primarily based on our experiences with consumer applications from the big web companies. The adoption of AI in the enterprise has been slow due to various challenges. Business applications address far more complex problems and the data needed to address them is less plentiful. There is also the critical need for alignment of AI with relevant business processes. In addition, the use of AI requires new engineering practices for application maintenance and trust.
So, how do you deliver working AI applications in the enterprise?
Beyond Algorithms: Delivering AI for Business answers this question. Written by three engineers with decades of experience in AI (and all the scars that come with that), this book explains what it takes to define, manage, engineer, and deliver end-to-end AI applications that work. This book presents:
- Core conceptual differences between AI and traditional business applications
- A new methodology that helps to prioritise AI projects and manage risks
- Practical case studies and examples with a focus on business impact and solution delivery
- Technical Deep Dives and Thought Experiments designed to challenge your brain and destroy your weekends
Author(s): James Luke, David Porter, Padmanabhan Santhanam
Publisher: CRC Press/Chapman & Hall
Year: 2022
Language: English
Pages: 300
City: Boca Raton
Cover
Half Title
Title Page
Copyright Page
Table of Contents
Authors
Acknowledgements
PROLOGUE
Chapter 1 Why This Book?
AI IS EVERYWHERE
ENTERPRISE APPLICATIONS
AI WINTERS
WHAT IS DIFFERENT NOW?
PROCEED WITH CAUTION!
DELIVERING AI SOLUTIONS
BETTER UNDERSTANDING OF AI IS CRITICAL FOR SOCIETY
TARGET AUDIENCE FOR THE BOOK
AN OUTLINE OF THE BOOK
REFERENCES
Chapter 2 Building Applications
WHAT’S DIFFERENT ABOUT AI WHEN BUILDING AN APPLICATION?
PROMINENT AI APPLICATIONS OF THE LAST SEVEN DECADES
AI OR NO AI?
THE PRESENT – THE DOMINANCE OF THE WEB
THE FUTURE – THE ENTERPRISE STRIKES BACK
EXAMPLES OF REAL ENTERPRISE APPLICATIONS
WHERE DO YOU INTRODUCE AI?
ACTIVITIES IN CREATING AN AI APPLICATION
COMPLEXITY OF AI APPLICATIONS
ARCHITECTURAL AND ENGINEERING CONSIDERATIONS
THREE STAGES OF AN ENTERPRISE AI APPLICATION
ENABLING ENTERPRISE SOLUTIONS AT SCALE
IN SUMMARY – ARE YOU READY TO START BUILDING APPLICATIONS?
REFERENCES
Chapter 3 It’s Not Just the Algorithms, Really!
INTRODUCING ALGORITHMS
ALGORITHMS IN AI
ALGORITHM ADDICTION
APPLICATIONS VERSUS THE UNDERLYING TECHNOLOGY
ALGORITHMS AND MODELS
OBJECT DROPPING PROBLEM
UNDERSTANDING THE OBJECT DROPPING DATA
FOUR MODELS TO PREDICT OBJECT BREAKAGE
COMPARING THE TWO ML APPROACHES
COMPARING PHYSICS MODEL WITH ML
WHAT ARE THE ML ALGORITHMS ACTUALLY LEARNING?
FEATURE DEFINITION AND EXTRACTION
REVENGE OF THE ARTIFICIAL NEURAL NETWORKS
HUMAN INTERPRETATION OF ARTIFICIAL NEURAL NETWORKS
SO WHICH ALGORITHM IS BEST?
TRANSFER LEARNING
REINFORCEMENT LEARNING
BRAIN VERSUS ARTIFICIAL NEURAL NETWORKS
FUNDAMENTAL PRINCIPLES AND FUNDAMENTAL MISTAKES
SO … IT REALLY ISN’T ALL ABOUT THE ALGORITHM
IN SUMMARY – THERE REALLY IS SO MUCH MORE TO AI THAN THE ALGORITHMS
REFERENCES
Chapter 4 Know Where to Start – Select the Right Project
THE DOABILITY METHOD
INNOVATION AND EMERGING TECHNOLOGIES
A PORTFOLIO-BASED APPROACH
DOABILITY METHOD STEP 1 – TO AI OR NOT AI
THREE RECOMMENDATIONS FROM DOABILITY METHOD STEP 1
DOABILITY METHOD STEP 1 – WORKED EXAMPLES
DOABILITY METHOD STEP 2 – PRIORITISING AI PROJECTS IN THE PORTFOLIO
IN SUMMARY – SUCCESS OR FAILURE WILL DEPEND ON SELECTING THE RIGHT PROJECT
REFERENCES
Chapter 5 Business Value and Impact
WHAT IS DIFFERENT ABOUT AI APPLICATIONS?
BUILDING BUSINESS CASES
STAKEHOLDERS
MEASURABILITY AND UNDERSTANDABILITY
IMPORTANCE OF ETHICS IN AI DEVELOPMENT
DELIVERING TRUSTWORTHY AI
FAIRNESS AND BIAS
EXPLAINABILITY
TRANSPARENCY
TACKLING THE WEAKNESS OF ML SYSTEMS
IN SUMMARY – THERE’S MORE TO VALUE THAN MONETARY RETURN
REFERENCES
Chapter 6 Ensuring It Works – How Do You Know?
MANAGING QUALITY OF TRADITIONAL SOFTWARE
MANAGING QUALITY OF AI APPLICATIONS
STATISTICAL ACCURACY
COST FUNCTIONS
MULTIPLE OUTCOMES
QUALITY METRICS FOR NATURAL LANGUAGE UNDERSTANDING
WHAT DOES THIS MEAN IN PRACTICE?
HOW ACCURATE DOES IT NEED TO BE?
WHERE DO YOU ASSESS ACCURACY AND BUSINESS IMPACT?
OPERATING WITHIN LIMITS
QUALITY ATTRIBUTES OF TRUSTWORTHY AI SYSTEMS
IN SUMMARY – IF THE AI ISN’T TRUSTWORTHY, PEOPLE WON’T TRUST IT
REFERENCES
Chapter 7 It’s All about the Data
DATA TSUNAMI
DATA TYPES
DATA SOURCES FOR AI
DATA FOR THE ENTERPRISE
ENTERPRISE REALITY
HUMANS VERSUS AI – LEARNING AND DECISION-MAKING
DATA WRANGLING
HOW MUCH DATA DO WE NEED?
SO, WHAT FEATURES DO WE NEED?
ENABLING EXPANDING FEATURE SPACES
WHAT HAPPENS IN THE REAL WORLD?
COPING WITH MISSING DATA
USE OF SYNTHETIC DATA
MANAGING THE DATA WORKFLOW
IMPROVING DATA QUALITY
IN SUMMARY – IT REALLY IS ALL ABOUT THE DATA!
REFERENCES
Chapter 8 How Hard Can It Be?
DEMONSTRATIONS VERSUS BUSINESS APPLICATIONS
SETTING EXPECTATIONS … YOURS AND OTHERS!
DO WE NEED AN INVENTION?
CURRENT STATE OF AI
THE IMPORTANCE OF DOMAIN SPECIALISTS
BUSINESS CHANGE AND AI
AI IS SOFTWARE
THE GREAT REUSE CHALLENGE
THE AI FACTORY
IN SUMMARY – IT CAN BE AS HARD AS YOU MAKE IT
REFERENCES
Chapter 9 Getting Your Priorities Right
AI PROJECT ASSESSMENT CHECKLIST
USING THE DOABILITY MATRIX
IN SUMMARY – NEVER TAKE OFF WITHOUT COMPLETING YOUR CHECKLIST
REFERENCE
Chapter 10 Some (Not So) Boring Stuff
TRADITIONAL ENGINEERING
WHY IS ENGINEERING AI DIFFERENT?
FOUR PHASES OF AN AI PROJECT
DEVELOPING AN ENTERPRISE AI APPLICATION
AI MODEL LIFECYCLE
APPLICATION LIFECYCLE
APPLICATION INTEGRATION AND DEPLOYMENT
PROJECT MANAGEMENT
AUDITABILITY AND EXPLAINABILITY
SECURITY
IN SUMMARY – THE BORING STUFF ISN’T REALLY BORING
REFERENCES
Chapter 11 The Future
IT’S ALL ABOUT THE DATA – TRENDS IN THE ENTERPRISE
EFFICIENT COMPUTING FOR AI WORKLOADS – NEW PARADIGMS
ADVANCES IN ALGORITHMS – TARGETING DATA CHALLENGES AND NEURO-SYMBOLIC AI
AI ENGINEERING – EMERGENCE OF A NEW DISCIPLINE
HUMAN–MACHINE TEAMING
IN SUMMARY – SOME FINAL THOUGHTS
REFERENCES
EPILOGUE
INDEX
