aaai99

WHY I AM OPTIMISTIC: 

WHY I AM OPTIMISTIC Patrick H. Winston MIT Artificial Intelligence Laboratory

The salients: 

The salients Applications side: We have already won Science side: We are bound to win

The Applications: 

The Applications

We have expanded our frontiers: 

We have expanded our frontiers Lots of people Lots of good

A little good for a lot of people: 

A little good for a lot of people

A little good for a lot of people: 

A little good for a lot of people

A lot of good for a few people: 

A lot of good for a few people

A lot of good for a lot of people: 

A lot of good for a lot of people

We have exemplars of all kinds: 

We have exemplars of all kinds Large software companies Large entertainment companies Companies with huge IPOs Multidimensional multinationals A multitude of small companies

We were a one-horse field: 

We were a one-horse field Rule chaining Inheritance

Now we ride many horses: 

Now we ride many horses Rule chaining Generate and test Search Tree building Agents Neural nets Constraint propagation Inheritance Genetic algorithms Bayes nets Learning

And not just reasoning horses: 

And not just reasoning horses Vision Language and speech Infrastructure

We had a Pyrrhic victory: 

We had a Pyrrhic victory IO Memory Cables Power Tapes Disk Network

We learned negative lessons: 

We learned negative lessons Nobody cares about saving money Using cutting edge technology To replace expensive experts

We learned positive lessons: 

We learned positive lessons Everybody cares about New revenues Saving a mountain of money Increasing competitiveness

We changed the business model: 

We changed the business model Replaces Expensive People Saves Mountains Of Money Creates New Revenue Ferrets Blunder stoppers Novices Experts

The critic and the billionaire: 

The critic and the billionaire

What’s next: connections: 

What’s next: connections People Global Net Physical World Computers Enhanced Reality Intelligent Structures Useful robots Human Computer Interaction Information Access

The click-in phenomenon: 

The click-in phenomenon The fax machine The world wide web

The Science: 

The Science

Shrobe’s point: 

Shrobe’s point Applications drive science Unless they all look alike

Atkeson’s point: 

Atkeson’s point We could move to the center But, we might be kidding ourselves

My point: 

My point AI is applied computer science Much energy wonderfully used But consequently diverted

A 100 year enterprise: 

A 100 year enterprise Molecular Biology Artificial Intelligence 1950 2000 2050 1900

Why we are the way we are: 

Why we are the way we are Powerful Ideas Models of Thinking Reflection…Biology…Psychology

The standard paradigm: 

The Intelligent Reasoner Language Vision Input/Output Channels The standard paradigm

The dawn age: 

The dawn age

What went wrong?: 

What went wrong? We think with our eyes We think with our mouths We think with our hands Each faculty helps the others

What is the evidence?: 

What is the evidence? Armchair psychology Clues from the brain

Armchair psychology: 

Armchair psychology Hillis’s observation on the value of talking to yourself Everyone’s observation on the value of drawing a sketch.

From brain scanning: 

From brain scanning

Intelligence is in the I/O: 

Intelligence is in the I/O The Explanation Motor Reasoner Linguistic Reasoner Visual Reasoner

Language is first among equals: 

Language is first among equals Language reasons Language tells vision how to see Language tells motor how to act

Is it time to start over?: 

Is it time to start over? An I/O oriented paradigm Essentially free computation Important, inspiring allies

From brain rewiring: 

From brain rewiring

From watching infants: 

From watching infants

Is it time to start over?: 

Is it time to start over? An I/O oriented paradigm Essentially free computation Important, inspiring allies Accumulation of powerful ideas

Six powerful ideas: 

Six powerful ideas Recreated condition One-shot learning Memory is cheap Change matters Survival of the smallest Bi-directional search

Recreated condition: Minsky: 

Recreated condition: Minsky P / \ P P / \ P P / \ / \ P P P P / \ / \ / \ / \ P P P P P P P P / \ / \ / \ / \ / \ / \ / \ / \ P P P P P P P P P P P P P P P P *---------------------------------- | | K-line | *--------------> | *--------->

One-shot: Yip and Sussman: 

One-shot: Yip and Sussman ae p l Time Word Memory Rule Memory

One-shot: Yip and Sussman: 

One-shot: Yip and Sussman ae p l z Time Rule Memory Word Memory

One-shot: Yip and Sussman: 

One-shot: Yip and Sussman 100% Trials 500 Accuracy

Memory is cheap: Atkeson: 

Memory is cheap: Atkeson

Atkeson’s practice tables: 

Atkeson’s practice tables

Atkeson’s practice results: 

Atkeson’s practice results One stored trajectory Feedback only Three stored trajectories

Change matters: Borchardt: 

Change matters: Borchardt

Borchardt’s ladder diagrams: 

Borchardt’s ladder diagrams D A D A A A A Distance Speed Contact

Survival of the smallest: Kirby: 

Survival of the smallest: Kirby

Kirby’s phase transitions: 

Kirby’s phase transitions Time Coverage

Bi-directional search: Ullman: 

Bi-directional search: Ullman Model Image

Joyous inferences: 

Joyous inferences Powerful ideas Marvelous engineering Essential alliances

What about …: 

What about … Bayes and Markov Neural nets and connectionism Logic

WHAT WE MUST NOT DO: 

WHAT WE MUST NOT DO

Loose our faith: 

Loose our faith It will take 300 years All the low hanging fruit is gone We shouldn’t make predictions

Waste time arguing: 

Waste time arguing Is it possible? Is it successful? Is it really AI?

Squander our capital: 

Squander our capital One thousand people 10% interested in the science side 10% actually working on it 10% of the time

WHAT WESHOULD DO: 

WHAT WE SHOULD DO

Human Intelligence Enterprise: 

Human Intelligence Enterprise Vision, language, motor Free hardware Clues from the brain Powerful ideas Conceive and test models

Why we should do it: 

Why we should do it It can only be done once Revolutionary applications

What we should ask: 

What we should ask Why do we have discrete words? What do our inner agents say? How do they learn what to say? Do we see what chimps see? How did our faculties evolve? Why can’t we all play the piano?

So, here is why I’m optimistic: 

So, here is why I’m optimistic Nothing could possibly be more fun, exciting, rewarding, and glorious, than … Applications that really matter Figuring out our own intelligence