1 / 49

Toward Models of Surprise

Toward Models of Surprise. Eric Horvitz Microsoft Research & University of Washington. Surprise. Unexpected or umodeled future event or outcome of significant consequence . Unexpected event : “ We thought about it, but considered it would occur with probability < x ”

oro
Download Presentation

Toward Models of Surprise

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Toward Models of Surprise Eric Horvitz Microsoft Research & University of Washington

  2. Surprise Unexpected or umodeled future event or outcome of significant consequence. • Unexpected event: “We thought about it, but considered it would occur with probability < x” • Unmodeled event: “We never even thought about that…” • Significant consequence: Significant positive or negative swing in utility of outcome for observer or group • Events can be personal, local, national, world

  3. Surprise Unexpected or umodeled future event or outcome of significant consequence. • Surprise defined in terms of an observer’s expectations • Deliberation and modeling can influence degree of or nature of surprise • Moving from surprise to plausible expectation • Expectation of classes or properties of surprise should given that a surprise occurs.

  4. Forecasting Surprise • Qualitative brainstorming about likelihood and nature of surprises over future time periods from knowledge, trends, and historical data • Opportunity: Statistical models of surprise • Learn from databases of past surprise to predict future surprises • Opportunity: Statistical models of predictive competency • Context- and content-centric failure to predict with accuracy Qualitative and quantative approaches

  5. Last 12 months Top n Properties 2004 2003 2002 Top n Top n Properties Top n Properties Properties SurpriseCasting: Abstract and Project • Example: • Identify top n most surprising things in prior 1 year (world events or personal life) • Repeat for 5 years of surprises • Characterize nature and rate of significant surprises • Consider top n over longer-term periods in same manner

  6. Surprise Forecasting: 2005-2020 Surprise Types Examples Actions Attention & Information SurpriseCasting: Abstract and Project • Turn to the future with statistics about types, examples. • Consider triaging of attention and recommending actions. Assume type occurs; give top m examples per type Recommended triage of focus of attention: additional study, info. gathering, monitoring per type Recommended actions to react, contain, disrupt negative suprises and/or exploit positive surprises Top n surprise types, by class or properties Surprises in technical sophistication of weaponry available inexpensively and universally Largescale simultaneous & continuous disabling of communication and sigint satellites available to multiple states.

  7. Surprise Forecasting: 2005-2020 Surprise Types Examples Actions Attention & Information SurpriseCasting: Abstract and Project • Deliberation about potential surprises: “Backcasting” • Chaining from surprising outcome to explore feasibility Assume type occurs; give top m examples per type Recommended triage of focus of attention: additional study, info. gathering, monitoring per type Recommended actions to react, contain, disrupt negative suprises and/or exploit positive surprises Top n surprise types, by class or properties Exotic transnational actors with access to increasing technical prowess. Terrorist organization establishes secret alliance with seemingly friendly former adversary state, intolerant of US “empire”; gains access to nuclear arms with deniability for former adversary.

  8. Expecting Surprises • Technical developments • International political terrain

  9. Technical Surprises from Past • Big surprises over 50 years • Flight • From likely unachievable dream, to major pillar of force projection and deterrence in just a “few summers”… • Quickly melted into background of obviousness • In a few years: From awe of a “canvas flying” contraption to concerns about low-salt meal while streaking across ocean in widebody jet. • Nuclear fission and fusion • Theory of Relativity and implications out of the blue • Shifts in balance of power • Risk to nation, human civilization

  10. Classes of Future Technical Surprises? • Computation • Intelligent systems change world (consumer, defense, communications, etc.) in a fundamental way? • Energy • New surprise source of energy changes economics of world • Changes nature of powerplants in our and adversaries weapon systems • Shift in balance of power hinge on the new source • Asymmetric force have new tool on their side for unleashing chaos and disruption • Biology • Assembly of Polio virus via internet and mail order in 2002 heralds era of bio-hacking • Major disruptions via pandemics

  11. Futurist Community: “Wildcards” “A wild card is a future development or event with a relatively low probability of occurrence but a likely high impact on the conduct of business” BIPE Conseil, Copenhagen Institute for Futures Studies, Institute for the Future: Wild Cards: A Multinational Perspective, Institute for the Future 1992

  12. Futurist Community: “Wildcards” • Futurist discussion of wildcards, “futurequakes” • Are natural, human-caused, or combinations • Are negative and positive • Multiple wildcards can interact • Breaks, discontinuities vs. trends • Peterson (2000), Out of the Blue: How to Anticipate Big Future Surprises • Cornish (2003), The Wild Cards in Our Future • Steinmüller (2003), The Future as Wild Card: A Short Introduction to a New Concept

  13. On Expected Surprises… • Earth & Sky • Biomedical Developments • Geopolitical & Sociological changes • Technology & Infrastructure Upheaval • Surprise Attack • Spiritual & Paranormal J. Petersen, Out of the Blue: How to Anticipate Big Future Surprises, 2000.

  14. On Expected Surprises… We are now living in another period of significant transition--a foreshortened span of time, during which our surroundings and experiences will change more than during any era in history. Humanity has never lived through the convergence--and, in some cases, the collision--of global forces of such magnitude and diversity. … One foreseeable outcome might be global instability; another, a planetary renaissance. J. Petersen, Out of the Blue: How to Anticipate Big Future Surprises, 2000.

  15. On Expected Surprises… … In any case, during the next two decades, almost every aspect of life will be fundamentally reshaped. . J. Petersen, Out of the Blue: How to Anticipate Big Future Surprises, 2000.

  16. Technological Prowess and Wildcards Rapidly growing technological prowess of humans has, for the first time in history, produced new classes of wild cards that could potentially destroy the whole human race. The new kinds of wild cards are simply too destructive to allow to happen and therefore require active attempts to anticipate them and be proactive in dealing with their early indicators. "Peterson, Out of the Blue: How to Anticipate Big Future Surprises, 2000

  17. Preparing for Wildcards Wild cards can be anticipated and prepared for. Thinking about a wild card before it happens is important and valuable. The more that is known about a potential event, the less threatening it becomes and the more obvious the solutions seem. "Peterson, Out of the Blue: How to Anticipate Big Future Surprises, 2000

  18. Backcasting Cornish: Multipronged Aerial Terrorist Attack was identified, e.g.,1987 article (B. Jenkins); 1994 article (M. Cetron): "Targets such as the World Trade Center not only provide the requisite casualties but, because of their symbolic nature, provide more bang for the buck.” "In order to maximize their odds for success, terrorist groups will likely consider mounting multiple, simultaneous operations with the aim of overtaxing a government's ability to respond, as well as demonstrating their professionalism and reach." “Despite all this, terrorism will remain a back-burner issue for Western leaders as long as the violence strikes in distant lands and has little impact on their fortunes or those of their constituents.” E. Cornish,The Wild Cards in Our Future. The Futurist. Washington: Jul/Aug 2003.Vol. 37, Iss. 4.

  19. Terrorists are unlikely to have their own aircraft. Could they buy one? Perhaps, but the cost would be enormous and authorities might ask too many questions. Could they hijack a military bomber and load it with bombs? They would have to be awfully clever and lucky to get onto an air base, seize a plane, load it with bombs, and fly it to a target. Success would be unlikely, and terrorists would know it. Could they hijack a military bomber and load it with bombs? Could they steal a plane? • An individual might have difficulty intimidating passengers and crew, but a group could probably do it even with modest weaponry. Getting into the pilot's cabin seems possible. So taking command of the plane seems feasible. But would the pilot follow instructions? Some instructions, maybe, but perhaps not all-especially if asked to do something suicidal. Could a terrorist be trained as a pilot to take over? Yes; a number of men in terrorist organizations have been trained as pilots. But how can an aircraft cause destruction if it has no bombs to drop? It can simply crash into its target. Such a crash should wreck a building; a skyscraper would be especially vulnerable. Of course, everybody on the plane would be killed, but if the terrorists are ready to die themselves, the deaths of the passengers serve as an added benefit. Scores of terrorists have gone on similar suicide missions by strapping bombs to themselves. We can now envision a group of terrorists hijacking an airliner and crashing it into an enemy target. But what target might it be? If we had read Cetron's article, we would already have one target to think about-the World Trade Center. In fact, Islamic terrorists did attack the Center in 1993, but their car bomb failed to destroy the building. • However, Cetron suggested that multiple targets might be selected for an attack. So we might stress the possibility of a larger, wider attack involving more terrorists and multiple planes and targets. Backcasting How might we proceed to evaluate this warning? • Poll experts on likelihood and targets of aerial suicide attacks. Assume aerial attacks plausible. • Backcasting: Reason about the challenges and solutions.

  20. Research on Statistical Models of Surprise • Most reasoning about surprise qualitative • Challenge: Data sparcity • Pushing on theoretical foundations of surprise • Example: Research traffic prediction service, Seattle

  21. Streaming Intelligence Sensing Learning & reasoning Web service Rich sensing Portable device Continual sensing, learning, and reasoning always available in stream of daily life

  22. Time until jam Times until jams clear Predictions in Your Pocket

  23. Multiple views on traffic Weather Major events • Event store • Learning • Reasoning Incident reports Operator ID: Nick Heading: INCIDENT Message: INCIDENT INFORMATION Cleared 1637: I-405 SB JS I-90 ACC BLK RL CCTV 1623 – WSP, FIR ON SCENE Traffic Prediction Challenge

  24. Operator ID: Nick Heading: INCIDENT Message:INCIDENT INFORMATION Cleared 1637: I-405 SB JS I-90 ACC BLK RL CCTV 1623 – WSP, FIR ON SCENE From Incident Reports to Salience e.g., Accidents How many lanes? Emergency vehicles?

  25. Data Set: Data Abstraction Lane Station Bottleneck

  26. Bayesian network forecasting model • Base-level predictions Machine Learning Database of sensed events • Data store Max likely time until bottleneck evaporates Building Predictive Models from Data • System-wide • status & dynamics • Incident reports • Sporting events • Weather • Time of day • Day of week • Season • Holiday status

  27. Predict Future Surprises Surprise at time tin thefuture: Infer surprise with model built from training set Evidence Time of day, day of week Bottlenecks, flows and their durations Incident reports Weather Games Accidents

  28. Probability ! ! Database of surprising events Future traffic • Data store • System-wide • status & dynamics • Incident reports • Sporting events • Weather • Time of day • Day of week • Season • Holiday status Human Forecaster Real World Outcome Building Models of Surprise • Learning predictive model of surprise • Model of user expectation and suprise • Models of future surprises Expectations Events at t • Major events • Weather • Time of day • Day of week • Holiday status

  29. Probability ! ! Database of surprising events Future traffic • Data store Expectations Machine learning • Major events • Weather • Time of day • Day of week • Holiday status • System-wide • status & dynamics • Incident reports • Sporting events • Weather • Time of day • Day of week • Season • Holiday status Human Forecaster Real World Outcome Building Models of Future Surprises • Learning predictive model of surprise • Model of user expectation and suprise • Models of future surprises Events at t Events at t-T • System-wide • status & dynamics • Incident reports • Sporting events • Weather • Time of day • Day of week • Season • Holiday status

  30. Base-level predictions • Context-sensitive error models • Data store • Inference • User logs • Surprise forecasting models Surprise & surprise forecasting Adding Surprise Modeling Traffic forecasting service • System-wide • status & dynamics • Incident reports • Sporting events • Weather • Time of day • Day of week • Season • Holiday status Competency annotation

  31. Where will there be surprises in time t?

  32. Sample Models and Exploration Influence of Seattle sporting events

  33. Sample Models and Exploration Influence of weather

  34. Sample Models and Exploration Influence of an accident at Bottleneck 15

  35. Anomaly at B4 Acc 15  .5 surprise -low Examining Details Acc 3  .5 surprise -.25 low -.25 high Influence of accidents at B15 and B3 on anomaly at B4 at 30 minutes

  36. Alerting on Device and Desktop • Specify preferences and download to device • Time- and route-specific preferences • Routes composed from bottlenecks • Multiple alert types

  37. Models of Surprise • Definitions of surprise • Qualitative and quantitative • Abstraction about past surprises to expect classes of surprise • Wild card notion in world of futurists • Directions: Statistical models of surprise

  38. Surprise Forecasting: 2005-2020 Surprise Types Examples Actions Attention & Information SurpriseCast Assume type occurs; give top m examples per type Recommended triage of focus of attention: additional study, info. gathering, monitoring per type Recommended actions to react, contain, disrupt negative suprises and/or exploit positive surprises Top n surprise types, by class or properties

More Related