Category: Games
Solve for the wartime presentation equilibrium
Ukrainian officials have run more than 8,600 facial recognition searches on dead or captured Russian soldiers in the 50 days since Moscow’s invasion began, using the scans to identify bodies and contact hundreds of their families in what may be one of the most gruesome applications of the technology to date.
The country’s IT Army, a volunteer force of hackers and activists that takes its direction from the Ukrainian government, says it has used those identifications to inform the families of the deaths of 582 Russians, including by sending them photos of the abandoned corpses.
The Ukrainians champion the use of face-scanning software from the U.S. tech firm Clearview AI as a brutal but effective way to stir up dissent inside Russia, discourage other fighters and hasten an end to a devastating war.
Here is the full story. Maybe this feels gruesome, but I am not sure we should let ourselves be led by the nose of our intuitions here. Furthermore, we have zero information on its effectiveness, or lack thereof. So I am not ready to have an opinion on this practice. We all seem fine with the idea of killing, so squeamishness on the “presentation side” probably is undertheorized.
I am more interested in what the next step looks like. If this stands a chance of being effective, how might you try to “improve” the presentation? Record death screams and send them in audio files? A virtual reality version? A “director’s cut” for the more committed audience members?
How about AI that scans the battlefield for fights your preferred side seems to be winning? Then do face scans of the opposing soldiers and using internet, text, or phone calls, invite their relatives to watch the struggle. Wouldn’t a fair number of family members click on that link?
Might some people crowdsource funding for extra footage, or shoot it themselves? I read this (New Yorker) report about the recent Brooklyn terror attack:
Many [bystanders] also responded as no one should ever do in an active-shooter scenario—when presented with an escape route, they instead stopped to record videos.
A yet more advanced version of the footage could throw in deep fakes of some kind? CGI?
Do you find this all more repulsive yet? Ever watched a war movie? We seem to accept those in full stride. It would be weird — but perhaps a coherent view nonetheless — to think “killing fine, phony movie of killing fine, movie of real killing just terrible.”
What do you all think?
For the initial pointer I thank Maxwell.
Under-signaling
Treasury Secretary Janet Yellen said the Biden administration would be prepared to use all its sanctions tools against China if Beijing moved aggressively toward Taiwan.
“I believe we’ve shown we can” impose significant pain on aggressive countries, as evidenced by sanctions against Russia, Yellen told lawmakers Wednesday as she testified before the House Financial Services Committee. “I think you should not doubt our ability and resolve to do the same in other situations.”
Here is the full Bloomberg story. If I were Xi Jinping, I would be heartened and encouraged by that ultimately rather lukewarm threat.
The game theory of attacking nuclear power plants
From my latest Bloomberg column (do I really need to indent my own text?):
“Putin would like to find a way of making nuclear threats without quite incurring the liability from … making nuclear threats.
Enter nuclear power plants. When Russian forces attack the plant, there is some chance that something goes wrong, such as a radiation spill. But more likely than not, the plant will hold up, and most dangerous processes can be shut down and the very worst outcomes avoided. You can think of Putin as choosing a “nuclear radiation deployment” with only some small probability.
Why might he do this? Well, he is showing that the use of broader nuclear deployments is not out of the question. He is also showing that he is willing to take a huge risk.
Most of all, he doesn’t much have to fear retaliation. The Western powers cannot know if these nuclear attacks are deliberate strategy or simply an accident of tactics in the field, and so — if only for that reason — they will not respond with a major escalation. If Russian forces moved on Estonia, they might be courting a very serious NATO response. But not in this situation.
You don’t have to believe that Putin sat in his lair rubbing his hands as he dreamed up this diabolical strategy. It’s also possible that the attack on the nuclear power plant started by mistake, or was ordered by lower-level commanders. Putin then simply allowed it to continue, perhaps out of a general love of chaos. At the very least, he did not consider it a priority to stop the attack.
Game theory doesn’t always have to be about explicit plans and intentions. It also can help explain why “invisible hand” mechanisms lead people to a particular point in the strategy tree, as if they had those strategies as conscious intentions.
Attacking the nuclear power plant also illuminates some other parts of game theory. Ukraine and its people are taking very heavy losses and are hoping for NATO to intervene on their behalf. If the conflict seems riskier to all of Europe, and not just Ukraine, the odds of such intervention improve.
In this sense, the attack on the nuclear power plant does not have to be entirely bad for Ukrainian prospects in the war. The Ukrainian leadership is rightly horrified by this attack, due to the risks for Ukrainian citizens. But the attack could also mobilize European public opinion on behalf of military intervention for Ukraine. If the war greatly increases chances for the spread of dangerous nuclear radiation, then the likelihood that Germany, France, Turkey and other nations will intervene also greatly increases.
Notice, however, that the Russian position here may be sounder than it at first appears. European citizens care more about radiation in Ukraine than do American citizens, for reasons of simple proximity. Putin may realize he can put Europeans at greater risk so long as he doesn’t provoke an intervention from the U.S. military, which would probably be decisive. It is a risky strategy that he might just get away with.
If you are the Ukrainian government, your incentive is to make the nuclear power plant attack sound as risky and precarious as possible. Indeed, Ukrainian President Volodymyr Zelenskiy has done exactly that.”
I am sorry to say that the column does not have an especially optimistic ending.
Norway Chess cancels Alexander Grischuk
He is kicked out of a forthcoming tournament, even though he has been critical of the war in Ukraine. In case you haven’t already guessed, Grischuk is Russian in his citizenship. This stuff never stays very accurate or fair.
Here is more on Grischuk and other Russian players, a grim and deeply unfair story (in German, covers Nepo, Svidler, Dubov, others who have spoken out). To be clear, I am strongly in support of pulling all tournaments from Russia and Belarus, as FIDE has done. It is targeting the individuals that I object to.
Elsewhere, from an email:
The editors of the “Studies in the History of Philosophy” have decided not to pursue the project of publishing a thematic issue devoted to Russian religious philosophy.
Russian cats are now cancelled too.
I would say this: if you are working in the United States and are from Russia, your chance of a big promotion just went way down, no matter what your political views. They are not going to make Chekhov Captain of the Enterprise, not now at least.
As for the oligarchs, I am all in favor of initiating court cases against Russian law breakers, whether in domestic courts or at The Hague. And if those individuals are found guilty, and the process generates yacht seizure as the appropriate remedy, bring it on. But just taking the yachts without true due process? Nein, Danke.
How much expected surplus do we want Putin to have?
My Bloomberg column is on another topic altogether, starting with bank runs, but this part I can reframe in terms of principal-agent theory. We want to squeeze Putin so hard that he “cries Uncle”, yet without eliminating his surplus so much that he takes a lot of extra risk. Hard to achieve both of those ends at the same time! Here is one bit reflecting that dilemma:
For a point of contrast on how decentralized incentives operate on each side, consider the nuclear alert ordered by Putin on Sunday. The chance of Russian nuclear weapons being ordered into actual use is small. But Putin faces a dilemma as he attempts to manipulate the decentralized systems of the Russian military. If he gave an order for a nuclear strike on a Ukrainian city, would the Russian military obey it? Whoever did would know they could be liable for war crimes.
The outcomes here are impossible to forecast, but the uncertainty works in favor of the Ukrainians. If it became known that Putin ordered a nuclear strike and was ignored, for example, he would become the proverbial “paper tiger” rather quickly and might lose power altogether.
These decentralized mechanisms potentially shift the entire logic of the war. Russia has to win fairly quickly, or these and other forces will increasingly work against it. Ukraine thus can fight for a military stalemate, but Russia cannot. The Russian forces must take increasing levels of risk, even if those risks have what decision theorists call “negative expected value” — that is, they serve as desperate gambles and on average worsen the Russian situation.
Of course that makes the war increasingly dangerous, and not just for the Ukrainians. If Putin is afraid the forces in the field won’t always carry out his orders, for example, he may order the launch of 10 tactical nukes rather than just one.
As AK would say, “Have a nice day.”
A simple model of what Putin will do for an endgame
I would start with two observations:
1. Putin’s goals have turned out to be more expansive than many (though not I) expected.
2. There are increasing doubts about Putin’s rationality.
I’ll accept #1, which has been my view all along, but put aside #2 for the time being.
In my simple model, in addition to a partial restoration of the empire, Putin desires a fundamental disruption to the EU and NATO. And much of Ukraine is not worth his ruling. As things currently stand, splitting Ukraine and taking the eastern half, while terrible for Ukraine (and for most of Russia as well), would not disrupt the EU and NATO. So when Putin is done doing that, he will attack and take a slice of territory to the north. It could be eastern Estonia, or it could relate to the Suwalki corridor, but in any case the act will be a larger challenge to the West because of explicit treaty commitments. Then he will see if we are willing to fight a war to get it back.
There are fixed costs to mobilization and incurring potential public wrath over the war, so as a leader you might as well “get the most out of it.” Our best hope is that the current Russian operations in Ukraine go sufficiently poorly that it does not come to this.
Addendum: And some good questions from Rob Lee.
The gender equality paradox seems to hold for chess
The gender-equality paradox refers to the puzzling finding that societies with more gender equality demonstrate larger gender differences across a range of phenomena, most notably in the proportion of women who pursue degrees in science, technology, engineering, and math. The present investigation demonstrates across two different measures of gender equality that this paradox extends to chess participation (N = 803,485 across 160 countries; age range: 3–100 years), specifically that women participate more often in countries with less gender equality. Previous explanations for the paradox fail to account for this finding. Instead, consistent with the notion that gender equality reflects a generational shift, mediation analyses suggest that the gender-equality paradox in chess is driven by the greater participation of younger players in countries with less gender equality. A curvilinear effect of gender equality on the participation of female players was also found, demonstrating that gender differences in chess participation are largest at the highest and lowest ends of the gender-equality spectrum.
Here is the paper by Allon Vishkin, via @autismcrisis.
Chess sentences to ponder
Using a recent sample of super-elite chess games, the present paper shows that males are substantially quicker to quit when dominated by a female than by a male. In contrast, female players’ behaviour differs little as a function of the gender of opponents.
Here is the paper, via the excellent Kevin Lewis.
Solve for the Eastern equilibrium
Russia’s sabre-rattling in Ukraine has reignited the debate in Finland as to whether the Nordic country should join Nato, defying demands from Moscow that seek to limit expansion of the military alliance in Europe.
Both president Sauli Niinisto and prime minister Sanna Marin used their new year addresses to underscore that Finland retained the option of seeking Nato membership at any time.
“Let it be stated once again: Finland’s room to manoeuvre and freedom of choice also include the possibility of military alignment and of applying for Nato membership, should we ourselves so decide,” Niinisto said.
Here is more from Richard Milne at the FT.
Some simple game theory of Omicron
Let’s say that everyone is totally reckless, and they go to Christmas Eve “Omicron parties.” A week or two from now the virus has cleared their systems and I, who stay at home and blog, can then go out and frolic. Even if they stay sick, or if they die, they are removed as sources of potential infections for others (see below for new variants, possibly from the immunocompromised).
If I know that is happening, I find it easy to stay at home for a week. I look forward to my pending freedom. In other words, right now my behavior becomes safer. I engage in intertemporal substitution.
Alternately, let’s say that quite a few people decide to behave more safely. They stay at home and avoid the Omicron parties, and furthermore they go about with a mask in Whole Foods and don’t go to bars at all. The Omicron pandemic, instead of being over in two weeks, can run on for months, depending on the exact numbers of course. There is a ready stock of “not yet infected with Omicron” potential victims to keep the virus circulating. And that means ongoing risk for me.
Returning to my decision calculus, I can wait a week but I cannot stay at home for a month or two. So I know I am going to go out, and I expect I am going to get Omicron. So I might as well go out now. My behavior becomes riskier.
Get the picture? If one set of people behave more safely, another set takes more risks. And vice versa.
This is one reason why moral exhortation, or for that matter policy interventions, may be less than effective in our current moment.
It is also a reason why telling people “don’t worry about it!” doesn’t fully translate at the collective level either.
Of course you can modify these scenarios with reinfection risk, new variants, and other factors.
And now it is over…
With both the Beatles and chess peaking this year in terms of media coverage, at times I have felt like I am thirteen years old again. But now the WCC match is over, and Magnus Carlsen has solidified his claim to GOAT. Carlsen has now won five such matches, and he has always won when he has needed to. Since he broke through the 2800 rating point, he has never fallen below it, not once. As a study in “management,” he is most of all a study in consistency. Nepo played even with him for five games, but then fell apart. Carlsen does not fall apart. Karjakin and Caruana played even with him for a whole match, but when the pressure was on in the rapid tiebreaks guess who was reaching new peaks?
I suspect this last match means the death of the slow classical format for the world championship. The last three matches have been deadly dull. You can cite particular reasons for the lack of excitement, but the fundamental problem is that the players are too good and a very well played chess game is a clear draw. It is hard to see how that gets reversed. On top of that the match format encourages risk-aversion and openings such as the Petroff for Black. There is too much advance openings preparation.
A Carlsen-Firouzja rapid match is what I wish to see, and somehow I expect the market will oblige. To have a repeat of what we just witnessed — even if the challenger shows up as the inspired player — just isn’t going to cut it. The cost is that we may not have a well-defined world champion by the time the next cycle moves toward its climax.
So, as of today, I predict that chess fundamentally has changed and won’t go back. No more Capablanca vs. Alekhine or Fischer vs. Spassky at slow speed. That’s just going to mean too many drawish opening choices.
Addendum: Please put aside your barbaric talk about Fischer Random 960. It obliterates the ability of the viewer to make sense of the board, so why bother? The rapid matches sponsored by Carlsen and others already have shown there are simpler, more viewer-friendly, and more intuitive ways to restore excitement to the games.
Lessons for remote work, from professional chess
During the COVID-19 pandemic, traditional (offline) chess tournaments were prohibited and instead held online. We exploit this unique setting to assess the impact of remote–work policies on the cognitive performance of individuals. Using the artificial intelligence embodied in a powerful chess engine to assess the quality of chess moves and associated errors, we find a statistically and economically significant decrease in performance when an individual competes remotely versus offline in a face-to-face setting. The effect size decreases over time, suggesting an adaptation to the new remote setting.
Here is the Economic Journal paper by . Via tekl.
Simple advice for watching and understanding on-line chess
Yes, the computer evaluations are extremely useful. But they are measuring the quality of the position when two computers are playing. Yet most of the games you care about tend to be two humans playing each other. And those humans do not play like computers. The computer might say the game is even, and maybe it is with perfect play, but one side can be much harder (easier) to play than the other. So I suggest this trick. Go to analysis.sesse.net, which covers top games (only). Scan down the vertical list of all possible moves and consider the distribution of outcomes. If the top move is great for White, but all the others are not, robustness is low, especially if the top move for White is not super-obvious (such as recapturing a Queen, etc.). If all the sequences look very good for White (Black), you will know that for humans the position probably is somewhat better for White than the single computer evaluation number will indicate. Robustness against human error will be present.
For the Carlsen match, here is a good Twitch stream, currently with Caruana as commentator.
Carlsen vs. Nepo
Here is my Bloomberg column on that topic:
He [Carlsen] recently opined that he is lucky to be facing Nepo rather than two other potential challengers, Fabio Caruana or Ding Liren. That’s the kind of trash talk most sports competitors frown upon for fear of motivating opponents.
Carlsen also has been engaging in online marathons of “bullet chess,” exactly the kind of attention-disrupting, energy-draining stunt contenders are supposed to avoid. In a bullet game, each player has only one minute for all the moves. The pace is so rapid the games are hard to watch, much less play. Carlsen also made a recent appearance in Dortmund, Germany, in part to pose for a photo with a Norwegian soccer player. Nepo, in contrast, claims to have done an “insane amount of work” for the event.
Will the fast thinking of bullet chess help Carlsen see more moves during the much slower time controls of the match with Nepo? (A championship game can easily last four hours or more.) Or maybe the bullet success will intimidate Nepo?
Carlsen also is making it clear that for him, chess is a business proposition. His parents set up a company in his name when he was 16, and the commercial empire since has expanded. Carlsen has worked as a fashion model, endorsed an online sports betting site, and worked with a Norwegian water company. He sponsors a leading chess app and has organized his own series of online chess tournaments, played with more rapid time controls, during the pandemic. Those events arguably have attracted more attention than any of the mainstream tournaments.
Carlsen is probably at the point where even a loss in the match would barely affect his income stream, and that is a dangerous motivational place to be…
Nepo is considered a super-talented but inconsistent player, one who does not bounce back well from adversity. But if he stays focused he could pose a formidable challenge. He was never expected to be a challenger in the first place, so he may feel he has little to lose and, in accord with his naturally aggressive style, he can take all the chances he wants. Carlsen is considered the superior player, perhaps the greatest ever, and remains a heavy favorite with the sports betting sites.
I am picking Carlsen to win. And on the future of chess:
Carlsen has argued that the mainstream matches of classical chess are too slow and yield too many draws. He would prefer a time limit of around 25 minutes per game per player to become the default. Why shouldn’t the world of chess switch over to a system that spectators seem to prefer?
If Carlsen retains his title, he may well lead such a switch, and it would be hard for the chess establishment to resist. If Nepo wins the match, Carlsen might secede from the current system, causing the chess world to splinter.
What we are seeing in the lead-up to this match is this: A healthy chess world is going to be a more diversely organized chess world, with a lot of disagreement over which forms of chess are most important. Twitch streaming and YouTube already have joined the mix. Chess is likely to retain its recent popularity, but in doing so it will fully realize its destiny as the esport it has already become. The good news is that if you don’t like the outcome of the upcoming chess drama, you can find another one to watch the next day.
Recommended.
Where to put your nuclear arsenal
I’ve been thinking about the article on MAD you linked to: Haller & Fry’s “The Math is Bad”. Their point — that you have to run the game theory for the case where a surprise first launch has already occurred — is interesting.
I agree MAD looks bad in that scenario. But I think the authors misunderstand why. And therefore their proposed solution — harden & build more capability — won’t work.
From a MAD point of view it’s incredibly stupid to put all your Minuteman missiles in a vast empty area no one cares about. Obviously the better placement would be to intermix the missiles with major urban centers.
There’s a reason the Minutemen aren’t scattered about New York, Chicago, Los Angeles, Houston, et all, and it’s not because our 1950’s leaders were stupid. It’s because we’re the good guys (or at least we were), and the good guys are inherently at a disadvantage when it comes to fighting.
It is at least possible to imagine a US president, facing either confirmed missiles in the air or the immediate aftermath of a successful first strike on the minutemen, might ask themself, at least for a moment, “what would be best for (my grandchildren) humanity?” rather than resignedly push the red button whilst saying “even though this won’t help anything, MAD requires I now launch more missiles.”
From that perspective, it really doesn’t matter how formidable our second-strike capacity is. Our enemies will *always* question our willingness to launch a return strike (on no doubt much messier targets). Indeed, during the cold war, even the allegedly inhuman Soviets worried about the human element, creating and possibly even implementing the famous Doomsday machine referenced in Dr. Strangelove in an attempt to prevent some wishy-washy comrade from choosing, in the heat of the moment, to avoid exterminating all life on the planet.
That doesn’t mean MAD is invalid, however.
There is another important component to the deterrent that Haller and Fry don’t consider: it may be that use of nuclear weapons even with no return strike is still not a survivable event. Even if fallout/nuclear winter effects prove mild, a first strike on even the smallest scale would upend the world. There is no leadership in any nation (save possibly North Korea) that could reasonably expect to survive the consequent metaphorical fallout.
This was put a little more pithily in the 1995 film “Crimson Tide,” when Denzel Washington says to Gene Hackman, “In the nuclear world, the true enemy is war itself.”
That is from Andy Lewicky.