1 Assisted Cognition Henry Kautz Don Patterson, Nan LI Oren Etzioni, Dieter Fox University of Washington Department of Computer Science & Engineering.

Slides:



Advertisements
Similar presentations
Technology Enabled High-Touch Care Majd Alwan, Ph.D. Medical Automation Research Center University of Virginia Improving healthcare quality and efficiency.
Advertisements

Using Visuals and Work Systems to add Structure to the Environment.
Opportunity Knocks: A Community Navigation Aid Henry Kautz Don Patterson Dieter Fox Lin Liao University of Washington Computer Science & Engineering.
11 Changing Demographics (US Census Dept, 2005). 22.
Recognizing Human Activity from Sensor Data Henry Kautz University of Washington Computer Science & Engineering graduate students: Don Patterson, Lin Liao.
State Estimation and Kalman Filtering CS B659 Spring 2013 Kris Hauser.
UNIVERSAL SYSTEMS MODEL
Martin Wagner and Gudrun Klinker Augmented Reality Group Institut für Informatik Technische Universität München December 19, 2003.
FixtureFinder: Discovering the Existence of Electrical and Water Fixtures Vijay Srinivasan*, John Stankovic, Kamin Whitehouse University of Virginia *(Currently.
Display of Information for Time-Critical Decision Making Eric Horvitz Decision Theory Group Microsoft Research Redmond, Washington 98025
Assisted Cognition Henry Kautz University of Rochester Computer Science.
Oregon Division of DD Tools to Help with Career Planning and Talking about Employment DD Council Guide for Career Discovery Employment Guide to Planning.
1 Assisted Cognition Henry Kautz, Oren Etzioni, & Dieter Fox University of Washington Department of Computer Science & Engineering.
Mining Motion Sensor Data from Smartphones for Estimating Vehicle Motion Tamer Nadeem, PhD Department of Computer Science NSF Workshop on Large-Scale Traffic.
Design Research Techniques for Elders with Cognitive Decline: Examples from Intel’s Digital Health Group Jay Lundell, PhD Margaret Morris, PhD.
CS B553: A LGORITHMS FOR O PTIMIZATION AND L EARNING aka “Neural and Genetic Approaches to Artificial Intelligence” Spring 2011 Kris Hauser.
1 Assisted Cognition Henry Kautz, Oren Etzioni, Dieter Fox, Gaetano Borriello, Larry Arnstein University of Washington Department of Computer Science &
Mobility for All Can one size fit all?. Universal Access and the Universe of One “There is no such thing as the average person.” –Don Norman, The Design.
Distributed Microsystems Laboratory: Developing Microsystems that Make Sense Sensor Validation Techniques Sponsoring Agency: Center for Process Analytical.
A Practical Approach to Recognizing Physical Activities Jonathan Lester, Tanzeem Choudhury, and Gaetano Borriello In Proceedings of the Fourth International.
Lecture 13 Revision IMS Systems Analysis and Design.
Review Questions List and describe the purpose of the four phases of Systems Analysis. The preliminary investigation phase quickly determines whether or.
Hidden Process Models Rebecca Hutchinson Tom M. Mitchell Indrayana Rustandi October 4, 2006 Women in Machine Learning Workshop Carnegie Mellon University.
Assisted Cognition Henry Kautz 590 AI – Autumn 2001.
Robotics for Intelligent Environments
What is Assisted Cognition? Henry Kautz University of Washington Computer Science & Engineering.
CLever: Building Cognitive Levers to help people help themselves Center for LifeLong Learning & Design University of Colorado at Boulder MAPS - LifeLine.
Probabilistic Databases Amol Deshpande, University of Maryland.
Task analysis 1 © Copyright De Montfort University 1998 All Rights Reserved Task Analysis Preece et al Chapter 7.
Scalable Server Load Balancing Inside Data Centers Dana Butnariu Princeton University Computer Science Department July – September 2010 Joint work with.
Help or Hal? Smart Homes & Elderly Care. Smart Homes A smart home (sometimes referred to as a smart house or eHome) is one that has highly advanced automatic.
Expert Systems Infsy 540 Dr. Ocker. Expert Systems n computer systems which try to mimic human expertise n produce a decision that does not require judgment.
Personalized Medicine Research at the University of Rochester Henry Kautz Department of Computer Science.
Bayesian Filtering for Robot Localization
Requirements, cont. …and a word on Ethics. Project Part 1: Requirements Gather data using one or more techniques Learn about environment, users, tasks,
Distributed Cognitive Aid with Interactive Task Guidance Edmund F. LoPresti 1,4, Ned Kirsch 2, Debra Schreckenghost 3, Richard Simpson 4 1 AT Sciences,
6 am 11 am 5 pm Fig. 5: Population density estimates using the aggregated Markov chains. Colour scale represents people per km. Population Activity Estimation.
Extracting Places and Activities from GPS Traces Using Hierarchical Conditional Random Fields Yong-Joong Kim Dept. of Computer Science Yonsei.
Chapter 8 Prediction Algorithms for Smart Environments
DATA MINING : CLASSIFICATION. Classification : Definition  Classification is a supervised learning.  Uses training sets which has correct answers (class.
 Definitions  Goals of automation in pharmacy  Advantages/disadvantages of automation  Application of automation to the medication use process  Clinical.
Requirements II: Task Analysis. Objectives By the end of the class, you will be able to… Write detailed task descriptions to inform design. Create scenarios.
Robust Activity Recognition Henry Kautz University of Washington Computer Science & Engineering graduate students: Don Patterson, Lin Liao, Krzysztof Gajos,
Prescribing Errors in General Practice The PRACtICe Study (2012) GMC Investigating Prevalence and Causes.
Making Sense of Sensors Henry Kautz Department of Computer Science & Engineering University of Washington, Seattle, WA Funding for this research is provided.
© 2007 Tom Beckman Features:  Are autonomous software entities that act as a user’s assistant to perform discrete tasks, simplifying or completely automating.
Introduction- Project Management By Ctrl+C & Ctrl+V 1.
The Science of Prediction Location Intelligence Conference April 4, 2006 How Next Generation Traffic Services Will Impact Business Dr. Oliver Downs, Chief.
Inferring High-Level Behavior from Low-Level Sensors Don Peterson, Lin Liao, Dieter Fox, Henry Kautz Published in UBICOMP 2003 ICS 280.
Recognizing Activities of Daily Living from Sensor Data Henry Kautz Department of Computer Science University of Rochester.
January Smart Environments: Artificial Intelligence in the Home and Beyond Diane J. Cook
Introduction to the 2007 Workshop on Intelligent Systems for Assisted Cognition Henry Kautz University of Rochester Department of Computer Science.
Learning and Inferring Transportation Routines By: Lin Liao, Dieter Fox and Henry Kautz Best Paper award AAAI’04.
University of Windsor School of Computer Science Topics in Artificial Intelligence Fall 2008 Sept 11, 2008.
Understanding Users Cognition & Cognitive Frameworks
Agents that Reduce Work and Information Overload and Beyond Intelligent Interfaces Presented by Maulik Oza Department of Information and Computer Science.
Inferring High-Level Behavior from Low-Level Sensors Donald J. Patterson, Lin Liao, Dieter Fox, and Henry Kautz.
Adaptive Tracking in Distributed Wireless Sensor Networks Lizhi Yang, Chuan Feng, Jerzy W. Rozenblit, Haiyan Qiao The University of Arizona Electrical.
Why Can't A Computer Be More Like A Brain?. Outline Introduction Turning Test HTM ◦ A. Theory ◦ B. Applications & Limits Conclusion.
1 Chapter 18: Selection and training n Selection and Training: Last lines of defense in creating a safe and efficient system n Selection: Methods for selecting.
Project ACCESS Henry Kautz, Dieter Fox, Gaetano Boriello (UW CSE) Don Patterson (UW / UC Irvine) Kurt Johnson, Pat Brown, Mark Harniss (UW Rehabilitation.
Artificial Intelligence: Research and Collaborative Possibilities a presentation by: Dr. Ernest L. McDuffie, Assistant Professor Department of Computer.
Robots.
IEEE AI - BASED POWER SYSTEM TRANSIENT SECURITY ASSESSMENT Dr. Hossam Talaat Dept. of Electrical Power & Machines Faculty of Engineering - Ain Shams.
REU 2009 Computer Science and Engineering Department The University of Texas at Arlington Research Experiences for Undergraduates in Information Processing.
Learning and Inferring Transportation Routines Lin Liao, Don Patterson, Dieter Fox, Henry Kautz Department of Computer Science and Engineering University.
Assisted Cognition Systems Henry Kautz Department of Computer Science.
Scottish National Burden of Disease, Injuries and Risk Factors study:
A probabilistic approach to cognition
Thrust IC: Action Selection in Joint-Human-Robot Teams
Presentation transcript:

1 Assisted Cognition Henry Kautz Don Patterson, Nan LI Oren Etzioni, Dieter Fox University of Washington Department of Computer Science & Engineering

2 Cognition in Context Can often compensate for physical disabilities by change in environment Wheelchairs Redesigned appliances Cognitive competence also depends on environment Can you cook dinner, given a dead animal, a stone knife, and set of flints?

3 Social Context The context for cognition involves both the physical and social environments Stability & organization of physical environment may reduce cognitive load Other people (e.g. a spouse) can actively assist in problem solving How can I make coffee? Which way is home?

4 The $80 Billion Question Can we build computer systems that (like a caregiver) actively assist a person with Alzheimer’s perform the tasks of day-to-day living? Enhance quality of life Prolong aging in place Lessen burden on other caretakers Depression affects 20% of Alzheimer’s patients, but 50% of Alzheimer’s caregivers Crisis in demographics – shortage of caretakers

5 The Assisted Cognition Project University of Washington Computer Science & Engineering UW Medical Center Alzheimer’s Disease Research Center (ADRC) UW Institution on Aging Outside Collaborators: Intel Research – Seattle and Jones Farm OGI/OHSU Elite Care

6 Vision Understanding human behavior from low-level sensory data Using commonsense knowledge Learning individual user models Actively offering prompts and other forms of help as needed Alerting human caregivers when necessary Computer systems that improve the independence and safety of people suffering from cognitive limitations by…

7 Example: Activity Compass Help user move between home and community Walking, riding in a car, public transport Predicts where user is going Offers simple directions Detects potential problems Is user on the wrong bus? Is user wandering?

8 Example: ADL Prompter 1. Joe enters bathroom at 9:00 am. 2. He turns on water, and picks up toothbrush. 3. Nothing happens for 30 seconds. AC system recognizes “tooth brushing” activity has stalled. 4. Prompts Joe to pick up toothpaste. Joe does so and completes task. 5. Joe leaves bathroom with water still running. AC system gently encourages Joe to go back and turn it off.

9 Common Architecture

10

11 Technical Approach GSP equipped Palm monitors location & velocity, communicates with server Dynamic Bayesian Net determines current mode of transportation Learned Markov Model predicts most likely activity path – i.e., user trajectory through time and space Each segment is a different mode of transport History, time of day, appointment calendar, bus schedules Guide user along activity path

12 Minimalist User Interface

13 User Feedback User may deviate from predicted path because System is wrong – need to update model User is in error – confused, forgetful System may ask for user for confirmation “Tap if you’re okay” Balance cost of annoying user vs. probability that user is in danger

14 Deciding When to Intervene (Horvitz 98) G = prediction that help is needed

15 Gathering Data

16 Velocity Histogram

17 Dynamic Bayesian Net M S V T1 T2 M S V T1 T2 TT+1 Mode Speed Timing BusStop Velocity BB

18 Mode Prediction

19

20 Current Work Measure accuracy of Markov Model for predicting activity path Compare other approaches Employ Relational Markov Model Less training data Increased power Planning algorithms for “error correction” E.g., once user has missed bus, find new path to achieve same goal

21 ADL Prompter General approach: build a probabilistic model of Common user goals “Plans” (complex behaviors) that achieve those goals Including failure modes How simple behaviors are sensed Run model “backwards” to interpret sensed data

22 Badge SensorDoor SensorGPS Location Get out of bed Walk to bathroom Flush Walk to bedroom Get into bed Night bathroom run Get out of bed Walk to bathroom

23 Badge SensorDoor SensorGPS Location Get out of bed Walk to kitchen Get crackers Walk to bedroom Get into bed Night snack run

24 Badge SensorDoor SensorGPS Location Night bathroom run Night pattern Night snack run Sleep

25 Timing Constraints Walk to bedroom Get into bed < 10 min Night bathroom run active [9 pm – 7 am] Night wandering violation

26 Summary: ADL Prompter Commonsense knowledge base of “significant” behaviors Hierarchically organized Probabilistic at all levels Several parallel ongoing activities possible Absolute and relative timing constraints Probabilities “tuned” by machine learning techniques for individual users Failure modes – points of possible intervention

27 Conclusions Growing research area combining AI, ubiquitous computing, and assistive technology NIST, AAAI, Ubicomp Workshops RESNA Gerontechnology Key idea: Patient and computer as a problem-solving team

28 End

29 Technical Foundations Hidden Markov models Mathematical framework for describing processes with hidden state that must be inferred from observations Hierarchical plan networks Represents how a task can be broken down into subtasks Hierarchical hidden Markov models Key to climbing food-chain!

30 Key Issue How to go from noisy and incomplete sensor measurements to A meaningful description of what a person is doing “Trying to brush teeth” “Trying to get home” A decision by the system about whether or not to intervene … in a principled and scalable manner!

31 Interventions Framework allows AC system to predict when a “failure” is likely Different failures have different costs Wandering in bedroom Wandering outside Forgetting to take medicine Forgetting to flush Must avoid:

32 Advertisement UbiCog 2002 – Workshop on Ubiquitous Computing for Cognitive Aids September 29, 2002 Gothenberg, Sweden Part of UBICOMP-2002, the major ubiquitous computing conference Some space still available, Henry Kautz

33 green – GPS readings (10 sec), yellow – location estimation (probability distribution)

34

35 Creating the User Model Training Data: 20,000 GPS readings Predicting mode 98% accuracy (10 FCV) Predicting next mode transition 97% accuracy (10 FCV)