How artificial intelligence is changing economic theory

July 17, 2015 by Leah Burrows
David C. Parkes, the George F. Colony Professor and Area Dean of Computer Science. Credit: Eliza Grinnell/Harvard Paulson School

A century of economic theory assumed that, given their available options, humans would always make rational decisions. Economists even had a name for this construct: homo economicus, the economic man.

Have you ever met a human? We're not always the most rational bunch. More recent theory confronts that fact, taking into account the importance of psychology, societal influences and emotion in our decision-making.

So, are the theories that are predicated on extinct? David C. Parkes, the George F. Colony Professor and Area Dean of Computer Science at Harvard John A. Paulson School of Engineering and Applied Sciences, doesn't think so. Humans may not always make , but well-conceived algorithms do.

In a paper out today in the journal Science, Parkes and co-author Michael Wellman, of the University of Michigan, argue that rational models of economics can be applied to artificial intelligence (AI) and discuss the future of machina economicus.

At first glance, neoclassical economic theory and AI seem like strange bedfellows. Where and how do they overlap?

Parkes: The idea of rationality is a shared construct between AI and economics. When we frame questions in AI, we say: what are the objectives, what should be optimized and what do we know about the world we're in? The AI/economics interface has become quite fertile because there is a shared language of utility, probability, and reasoning about others.

Take, for example, the revelation principle in economics, which is a theory by which the design of economic institutions, such as markets, can be restricted to those where it is in the best interest of participants to truthfully reveal their utility functions. Today's Internet advertising systems, which are populated by artificial trading agents, are an operational version of this . Search engines are designing interfaces where advertisers reveal their budget constraints and goals, and these systems then provide the algorithms to fit those needs. You don't see many mechanisms like this in human societies but we may see them more and more in AI systems.

Where does current economic theory fall short in describing rational AI?

Machina economicus might better fit the typical economic theories of rational behavior, but we don't believe that the AI will be fully rational or have unbounded abilities to solve problems. At some point you hit the intractability limit—things we know cannot be solved optimally—and at that point, there will be questions about the right way to model deviations from truly rational behavior.

Poker is great example of a complicated reasoning problem: a lot of information is missing, you don't know the cards of the other players, you're uncertain about the card that will be dealt next and you're reasoning against another reasoning agent.

Recently, researchers developed an algorithm that effectively solves Heads Up Limit Texas Hold'em, applying game theory from economics. Researchers have developed an AI that has attained perfect rationality in this setting, and have done so using a number of general-purpose techniques. But this has come about after decades of research, and only for a restricted, two player version of poker.

But perfect rationality is not achievable in many complex real-world settings, and will almost surely remain so. In this light, machina economicus may need its own economic theories to usefully describe behavior and to use for the purpose of designing rules by which these agents interact.

Besides poker, what would a rational AI system do better than a human?

One of the more complicated things people do is buying and selling property. It's actually really hard to describe to your real estate agent what you're looking for. Your broker may have some ideas, but real feedback only comes when showing you properties. It's an inefficient system. AI researchers have been developing the idea that an AI would elicit your preferences initially by direct query. An AI can show you a comparison between two houses and ask which one you prefer. As you answer, the AI can build a model of your preferences, adaptively eliciting information until it can reason that it knows enough about your preferences that the AI would be able to go out and— if it was very successful—know what house to buy on your behalf. Even if it were only moderately successful, the AI would bring back a couple of options that did a good job of optimizing your preferences within the market.

How would humans interact with these rational machines?

One way a machine can understand you and your preferences is to observe how you act. There is an approach called inverse reinforcement learning—economists call it revealed preference—where if an AI sees the decisions you make every day, it can begin to understand something about you. What your trade offs are, how you spend your time, what you like to wear and when, who do you like to talk to, who do you not like to talk to, etc. By observing your behavior, the AI can begin to build a model of your preferences. Then, you can imagine over time the AI could start acting on those preferences and interposing itself—hopefully not in a creepy way.

Researchers have been developing techniques that look at your electronic information stream—your emails, voicemails, social media use—and learn about such things as your work environment, its hierarchy, who your manager is, and who reports to you. From there, the AI can decide which communications you actually need to see when. It can know when you're in a meeting and only interrupt you when someone important calls. It can do that based on modeling the value to you of information versus the cost of interruption. As the AI begins to do more for you, it can learn based on the choices you make. Of course, this presupposes that you're making rational choices on your own behalf, so your revealed preferences may not be the same as what your preferences should be.

What are the biggest challenges in building machina economicus?

The problems AIs will be solving, whether they are in a market or social context, are complex problems, especially where there are other participants in the system.

Optimal behavior will often depend on the behavior of others, making this quite different from reasoning about an environment where you are the only actor. If an AI is acting to buy, sell or exchange information or to set the price for something, it needs to reason about what other AIs are doing in the system as well.

Are there dangers in building something that is too rational?

Rationality can lead to unintended consequences. If you tell an AI car to get into the city as quickly as possible, it might run some lights because its optimizing and reasoning about the probability of getting caught versus getting to its destination quickly.

Analogies like this ring true to the stock market as well. At the moment, we're living in a time where the presence of fast, algorithmic trading algorithms is leading to concerns about the fairness and efficiency of the stock markets. However, AI can also make markets more efficient, by doing a better job of matching supply and demand and allocating resources to those who need them and better understanding preferences and societal considerations.

There are also important questions about how rapid progress in AI will affect the workplace and the broader economy, both in the U.S. and globally, and this is an area that economists and policy makers should be looking at and are looking at.

Explore further: Researchers examines the true state of artificial intelligence

More information: "Economic reasoning and artificial intelligence." Science 17 July 2015: DOI: 10.1126/science.aaa8403

Related Stories

Recommended for you

Historic solar flight marks first round-the-world journey

July 26, 2016

The world's first round-the-world flight to be powered solely by the sun's energy made history on Tuesday as it landed in Abu Dhabi, where it first took off on an epic 25,000-mile (40,000-kilometer) journey that began more ...

How to build a 1,000mph car (by the scientists behind it)

July 22, 2016

It was a staggering feat, a car that went faster than the speed of sound. On October 15 1997, Andy Green travelled across the Black Rock Desert, Nevada, in the Thrust SSC at 763.035 mph, or Mach 1.02. Two decades on, that ...

16 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

luke_w_bradley
not rated yet Jul 18, 2015
Good read, I think the convergence of AI and economic systems, possibly through blockchain related economic systems, is an an area we could see some astounding breakthroughs. One thing I'm not so sure of is that AI will have a level of rationality that transcends humans at all in its beginnings though. I think of it almost more as animal and instinctual, but capable of living in ecosystems of big data our minds just aren't made for. For a time, it may be more like a useful hound dog for humans, sniffing out possibly interesting things for us to inspect with our more rational minds.
alohim_haushulamee
1 / 5 (1) Jul 18, 2015
AI is yet very limited when compared with human mind and decisive capabilities related to comfort with expertise, so my economics is not based on logic but on wisdom.
See my invention of "Universal Automatic Comprehension, so to capture the wrongs in your article.
thingumbobesquire
not rated yet Jul 18, 2015
Virtually worthless view of economics. The issue for real economics is what is required to uplift the standard of living globally.
viko_mx
5 / 5 (1) Jul 18, 2015
What is the definition of artifitial inteligence? Do the author of this article makes a difference between a computer program executing set of algorithms created by human intelligence and device with own imagination and ability for self making a decisions?
correo
not rated yet Jul 18, 2015
Before it becomes a general term, I would suggest changing the name "Machina Economicus", That's a big monstrosity in Latin.
"Homo economicus" (economic human) it's OK in Latin since homo is masculine as well as economicus, but
"Machina economicus" (denoting economic machine) is NOT latin at all, since machina is feminine while economicus is masculine. Therefore it should be "MACHINA ECONOMICA" with the adjective "economica" in the same gender, feminine, than "machina".

Besides that, congratulations for your article.
fay
not rated yet Jul 19, 2015
//More recent economic theory confronts that fact, taking into account the importance of psychology, societal influences and emotion in our decision-making.//
that doesnt at all mean our decisions are irrational, it means they are rational but take into account some other factors that are hard or impossible to measure, but have impact on the decision maker nonetheless. For example, is it irrational to buy some phone that is objectively worse in performance than other phones for the same price, but the brand is deemed really cool by decision maker`s friends or the whole society? Absolutely not, thats perfectly rational decision.
Macrocompassion
not rated yet Jul 20, 2015
The performance of our social system is not determinate in terms of individual behavior, but is found when considering the average properties of the whole Macro-Economics Social System (MESS). Thus the concept of homo economicus or whatever you wish to name it is flawed.
EnsignFlandry
5 / 5 (1) Jul 20, 2015
Virtually worthless view of economics. The issue for real economics is what is required to uplift the standard of living globally.


Which would be based on facts and rational decision-making. We already know much of what is required. Its that pesky human will that gets in the way. Look at economies around the world that are growing, and those that are not.
Returners
not rated yet Jul 20, 2015
Virtually worthless view of economics. The issue for real economics is what is required to uplift the standard of living globally.


Higher taxes on the insanely wealthy, and more rational spending of government money to maintain infrastructure and healthcare.

I want the government to own and operate energy infrastructure, because it would be much more economically efficient, and because government employees are appointed and have salary caps, while corporate CEOs do not have salary caps and hoard wealth. We get to vote for who does the hiring and appointing in the government. We don't get to vote for who gets to be the money-hoarding CEO of the corporation.

Capitalism is nothing more than a legalized pyramid scheme, and only the 1% benefit from it.

ryggesogn2
1 / 5 (1) Jul 21, 2015
I want the government to own and operate energy infrastructure, because it would be much more economically efficient,


What BS!

Nationalized industries all over the world are grossly inefficient and waste resources.

"But Pemex is in danger of breaking down. ''Financially, we are falling,'' its director, Raúl Muñoz Leos, said in an interview. Nearly every peso of Pemex's profits goes to run the government of Mexico. The company, after paying taxes and royalties, actually lost $3.5 billion in in 2001. Without major restructuring or tens of billions of dollars in foreign investment, Mr. Muñoz Leos warned recently, ''We would face, in the short term, a collapse.'' "
http://www.nytime...ine.html

We don't get to vote for who gets to be the money-hoarding CEO of the corporation.


Yes, you do. Every day you don't make a purchase from that company and buy from a competitor is a vote.
barakn
3 / 5 (2) Jul 21, 2015
Wow, soggyring2, a 12-year old article. Gosh, that's convincing.
ryggesogn2
1 / 5 (1) Jul 21, 2015
Wow, soggyring2, a 12-year old article. Gosh, that's convincing.


Is this convincing that govt monopolies fail?

"Mexico has opened the oil and gas industry in a bid to end a decade-long slide in production."
http://www.reuter...20150312
ryggesogn2
1 / 5 (1) Jul 21, 2015
Or this:

"Last December's energy overhaul, whose implementing legislation is currently being debated in Congress, allows private players into Mexico's oil industry for the first time since it was nationalized in 1938.

The modernization measures being considered include re-launching Pemex's brand, improving the image of the company's gas stations and customer service, bringing new products to market and forging partnerships with leading companies.

Some measures have already been implemented, including a training program that will eventually reach more than 26,000 managers, supervisors and sales staff, Lozoya said."
http://latino.fox...etition/

"Pemex would benefit from competition," Moreira said at today's conference. "There is nothing more efficient to make a company more competitive than to have competition.""
http://www.bloomberg.com/
ryggesogn2
1 / 5 (1) Jul 21, 2015
Higher taxes on the insanely wealthy,


" Maybe the next Bjorn Borg won't feel compelled to move to Monaco now that Sweden plans to scrap a decades-old "wealth" tax that imposes levies on assets — not just on income. …The move, expected to be approved by parliament later this year, underscores the country's efforts to keep successful Swedes and their capital at home by changing its fabled but costly welfare state.

"It's not sustainable to keep taxes that radically diverge from other countries," Finance Minister Anders Borg, who is not related to the tennis great, told The Associated Press on Thursday. "Not if you want the money to stay in the country.""
"[T]he wealthiest Swedes have fled the country, including IKEA founder Ingvar Kamprad, No. 4 on Forbes magazine's list of the world's richest people. He lives in Switzerland. "
http://www.cato.o...alth-tax
ryggesogn2
1 / 5 (1) Jul 21, 2015
government money


What is 'government money'? The money a govt prints? Or the wealth it plunders from its serfs?

Look at economies around the world that are growing, and those that are not.


Only those that enable that pesky human will. All the economies with significant govt control are not growing.
Dug
not rated yet Jul 22, 2015
"Researchers have been developing techniques that look at your electronic information stream—your emails, voicemails, social media use..." No, not really. They have tried, but the results are decidedly crappy. This is not that different from Google searches - which are actually less performance definitive now than they were five years ago, before all this crap AI crap tried to figure you out and what you need. Mostly, it's in the way of basic efficiency and accuracy - and they are using it anyway.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.