artificial intelligence and the car - what does the future...
TRANSCRIPT
Artificial intelligence and the car - what does the future hold? Dr. Joanna Batstone Vice President, IBM Watson @jbatstone Oct 22, 2018
Advances in Artificial Intelligence
Advancing the foundations of one of the most fundamental technologies of the 21st century
@IBMWatson
The birth of artificial intelligence - 1956 Dartmouth Conference
John MacCarthy Marvin Minsky Claude Shannon Ray Solomonoff Alan Newell
“We propose that a 2 month, 10 man study of artificial intelligence be carried out…The study is to proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it… We think that a significant advance can be made in one or more of these problems if a carefully selected group of scientists work on it together for a summer.”
Herbert Simon Arthur Samuel Oliver Selfridge Nathaniel Rochester Tenchard More
What are Artificial Intelligence, Machine Learning and Cognitive Computing?
@IBMWatson
Data mining
Pattern recognition
Natural language processing
Ontologies Statistics
Robotics
Expert systems
Neuromorphic
COGNITIVE COMPUTING
MACHINE LEARNING
ARTIFICIAL INTELLIGENCE
Big Data
Perception
Robotic Process Automation
Classification
MACHINE ENHANCED DECISION MAKING
AUTONOMOUS DECISION MAKING
Game playing
@IBMWatson
AI/ML as the domain of the academic and researcher.
Few applications for everyday life.
AI/ML Foundations
Artificial general intelligence and robotics enables
computers to surpass the abilities of humans at most physical and mental tasks
AI/ML Tomorrow
AI/ML in widespread use to enhance everyday life,
usually applied to specific narrow problems or tasks.
AI/ML Today
1997 - Deep Blue vs. Kasparov
@IBMWatson
2005 – Stanley vs. The DARPA Challenge
@IBMWatson
2011- Watson vs. Jeopardy
@IBMWatson
2017- AlphaGo Zero vs. itself…
The evolution of AI
Narrow AI Initial
Value Creation
Broad AI
Disruptive and Pervasive
General AI Revolutionary
▼ We are here 2050 and beyond 2010 and earlier 2015
Every business and enterprise is embracing AI
AI Everywhere
Healthcare Finance Agriculture Government Education Energy Science Business solutions
Instrumented Planet
Environmental solutions Digital agriculture Connected cars Geospatial-temporal data and analytics Smart sensors
Deeper Insights
Data-centric systems Distributed Deep Learning Neuromorphic systems Quantum computing Homomorphic encryption Machine foresight Cognitive discovery
Engagement Reimagined
Human-machine collaboration New AI modalities Augmented reality Global trade logistics Blockchain for payments
Personalization at Scale
Personalized healthcare Micro-segmentation Personalized finance Targeted marketing Personalized learning Individualized solutions
What lies ahead?
12 IBM Research / © 2018 IBM Corporation
Research is at the heart of advancing broad AI
Signal Comprehension: From video and text to rich human perception
Learning and Reasoning: From scalable machine learning to making a case
Interaction: Understanding language, tone, emotion and context
“A green bird sitting on top of a bowl”
Brain-inspired neuromorphic systems
Cars & the automotive industry
Olli is the first vehicle to utilize the cloud-based cognitive computing capability to analyze and learn from high volumes of transportation data, produced by more than 30 sensors embedded throughout the vehicle. The platform leverages four Watson developer APIs -- Speech to Text, Natural Language Classifier, Entity Extraction and Text to Speech -- to enable seamless interactions between the vehicle and passengers.
Local Motors Debuts "Olli", the first self-driving vehicle to tap the power of IBM Watson
Experience expectations are based on previous best experiences
DO NOT DISTRIBUTE - CONFIDENTIAL & PROPRIETARY
Drivers will customize the service to their preferences
DO NOT DISTRIBUTE - CONFIDENTIAL & PROPRIETARY
American drivers average more than 17,600 minutes, or about 293 hours behind the wheel
annually.
Average time spent behind the wheel each year is equivalent to
seven 40-hour work weeks.
*AAA Foundation for Traffic Safety. 2016
88% of drivers use their phones while driving.
During an hour-long trip, drivers spend an average of 3.5-minutes using their
phones.
http://blog.zendrive.com/distracted-driving/ 2017 .
The average driver will spend sixteen hours & twenty minutes using their
phone while driving.
The next 5 years will be more important
than the last 100
Frictionless loan origination
• Approval in minutes or hours (10x faster than 5 years ago)
• Analytics and machine learning to improve customer experience
Automated claims processing
• Rapid payment of claims with transparency and instant visibility
• Machine learning & AI to identify fraud & abuse and fast track others
Optimized healthcare management
• Automation of routine tasks, such as patient & equipment scheduling
• Analytics & AI to minimize wasted healthcare resources
Today with digital automation, a small number of experts can create great experiences for customers at scale
© 2017 International Business Machines Corporation 22
3 major ways for AI to drive value
Voice of the Customer Allows clients to understand their customer’s sentiment and trends beyond what’s gathered from traditional surveys and call center interaction(s).
Customer Care Provides an omni-channel way for customers to have a new engagement model with clients business while avoiding contact center traffic.
Expert Assist Helps customer-facing stakeholders when they need more information to help solve a problem, answer a question or cross-sell a customer.
© 2017 International Business Machines Corporation 23
No apps, no buttons, no passwords, no more computer stuff…
No more admin…
Just users & services…
Voice as the UX Model is the most Natural User Interface
”Directions to 123 Hudson Street, Detroit Michigan” “I need directions to Brighton’s Varsity Soccer Field” “I need directions to Aunt Rosie’s house.” “How long will it take me to get home”
“Where can I get a coffee?” “I’m looking for a pharmacy” “What’s the highest rated Italian restaurant within 10 miles?” ”Will I make it there before it closes?” “What time is my next meeting?”
“Why is my check engine light on?” “Are my tires at the recommended pressure?” “How do I check my oil level?” “Tell me when I need to get my tires rotated” ”Schedule an appointment at my dealership for Saturday morning”
Watson Assistant: • Automotive domain expertise • Integration with calendar, map, etc. • Can be invoked proactively on your
behalf by real world object event changes
• Use Case Focus: • Conversational interactions & Speech • Commerce • Maintenance • Safety • Multi tasking
IBM Corporation
Owning the Customer by removing Friction, Questions & Choices
IBM Watson & Xevo Delivers Major Merchant Brands to General Motors In-Car Commerce Platform
Enables POI Requests & Interactions
Ask about a
restaurant, coffee shop or dry cleaner.
Get a curated
response based on your priorities
Where can I get a coffee?
Which side of the car is my gas cap on?
Enables Car Manual Questions & Interactions
Ask questions based on the vehicle car manual.
Get a groomed
response based on vehicle & it’s manual
data
When is my next meeting?
Understands your Calendar & its
Context
Ask when your next meeting is while driving.
Get calendar data based on what will happen while in en-
route.
What’s my tire pressure?
Show me the fastest way to get home.
Understands Vehicle Data
Ask about vehicle status,
either real time or historical.
Get vehicle data while driving or proactively
when an event demands your attention
Understands navigation
Navigation &
Conversation simplifies getting around
Integrate navigation into
the entire experience
Differentiating Use Cases
© 2017 International Business Machines Corporation 27
Autoglass Body Repair Visual Recognition
Results:
• >70% improvement in quote processing times by using AI analyses to determine repair costs
• Optimizes damage advisors’ time and skills by allowing employees to focus on more complex claims
• Boosts efficiency of the quote generation process by supporting human decision-making with AI analyses
27
© 2018 IBM Corporation
Industry: Automotive
BMW Group now features weather data provided by The Weather Company within their ConnectedDrive systems in vehicles across North America. Additional IoT-based connected auto services from BMW will provide realtime, contextual updates about route, traffic and vehicle status in order to enrich the driving experience and make recommendations to the driver.
BMW and IBM explore conversational interfaces and integrated Weather data for connected auto services
3 Keys to Successfully Implementing AI
29 © 2018 IBM Corporation
Organizational Readiness – expedites AI technologies at the
enterprise level
Data Readiness – shapes the implementation timeline
and end user experience
Value Readiness – defines business outcomes
Multi-Cloud
COLLECT - Make data simple & accessible
ORGANIZE - Create a trusted analytics foundation
ANALYZE - Scale insights on demand
AUTOMATE - Apply ML everywhere
TRUST - Achieve trust & transparency
Make your data ready for AI
Protect your insights
Reimagine your workflows
Modernize for simplicity & agility
Data of every type, regardless of where it lives
Trust and transparency is the next step in the ladder to AI
30
AI bias will explode. But only the unbiased AI will survive.
Fair AI: Recognizing & Avoiding Biases
Data Bias
Model Bias
Re-training Bias
Training
Continuous reinforcement
Data
Deliver Trust & Transparency for Enterprise AI IBM AI OpenScale
35 35
• Automatically detects fairness issues at runtime
• Intelligently delivers bias mitigation help
• Provides traceability & auditability of AI predictions made in production applications
• Tracks AI accuracy in applications
• Explains an outcome/recommendation in business terms
IBM Cloud / Watson and Cloud Platform / © 2018 IBM Corporation
Explain how AI arrived at a prediction • Uses contrastive techniques to
explain model behavior in the vicinity of the target data point.
• Identifies feature weighting of most and least important features
• Displays factors that influence a prediction in simple terms.
Explainability
IBM Cloud / Watson and Cloud Platform / © 2018 IBM Corporation
Compassion Intuition Design
Value judgments Common sense
Compassion Intuition Design
Value judgments Common sense
Deep Learning Discovery Large-scale math Fact checking
Human Machine +
#DSES
THANK YOU
https://www.ibm.com/watson
@jbatstone