Games

Christina, an apparent MR reader, asked me whether it is really true that AI helps military defense more than military offense, as was previously argued by Eric Schmidt.  I can think of a few parallel cases:

1. In chess, AI clearly has helped the defense.  Top computer programs never play 32-move brilliant sacrifice victories against each other, a’ la Mikhail Tal.  Most games are drawn, and a victory tends to be long and protracted.  (Do note it is sometimes better to get the war over with and lose right away.)

2. In the NBA, analytics have helped offense more, for instance by showing that more attempted shots should be three-pointers.  Analytics of course is not AI, but you can consider it a more primitive form of using information technology to improve decisions.

3. It is interesting to ponder the differences between chess and the NBA as potential analogies.  In chess, the attack often “plays itself,” as the player with the initiative may be following fairly standard strategies of bringing the Queen and some lesser pieces in the neighborhood of the opposing King, or maybe just capturing material.  Finding the correct defense is often a more complex matter, and the higher quality of the chess-playing programs thus boosts defense more than offense.  Besides, under perfect information chess is almost certainly a draw, and the use of AI asymptotically approaches that outcome.

In professional basketball, the offense typically has more options and permutations, and given any offensive decisions, the defense often respond in fairly typical fashion, such as lunging at the player attempting a shot, or doubling Stephen Curry as he crosses the half-court line.  In those cases where the defense has more options, however, analytics conceivably could help basketball defense more than offense.  A (hypothetical) example of this would be using game tape and AI to see which kinds of tugs on the jersey best disrupt the shot or rhythm of the team’s leading scorer.  That said, most of the action seems to be in honing the options for the offense.

4. Is warfare more like chess or more like the NBA?

I believe the USA has more options in most of its conflicts, and thus AI will help the United States, at least at first.

In the Second World War the Nazis had more options than their opponents.  In the Civil War and American Revolution, however, the available offense was more static and predictable, and AI for those fighting forces might have helped the defense more.  In the Iran-Iraq war I suspect the defense had more options too.  Terror groups have more meaningful options than the forces defending against terror, and thus AI might help terror groups more than the defense, at least provided they had equal access to the data and to the technology (which is doubtful at this point, still as part of the exercise this is useful).

5. One important qualifier is that the chess and NBA examples already assume a game is on to be played.  A war, in contrast, is started as a matter of volition on at least one side.  If AI creates a new arms race of sorts, where one side at times opens up a decisive lead, that may provoke more decisions to engage and thus attack.  The mere fact that AI increases the variance in the power gap between the two sides may increase the number of attacks and thus wars.

So there is more to this question than meets the eye at first, and I have only begun to engage with it.

Addendum: AI is also spreading in the legal world, will this help defendants or plaintiffs more?

Cyber weapons are different. If you are a state and you let potential enemies know about your arsenal of cyber attacks, you are giving them the opportunity to fix their information systems so that they can neutralize the threat. This means that it is very hard to use cyber weapons to make credible threats against other states. As soon as you have made a credibly-specific threat, you have likely given your target enough information to figure out the vulnerability that you want to exploit.

This means that offensive cyber weapons are better for gathering intelligence or actually taking out military targets than for making threats. In this regard, they are the opposite of nuclear weapons, which are more useful as threats than as battlefield options. Nuclear weapons can create stability because they deter attacks. In effect, they create a stable system of beliefs where no state wants to seriously attack a nuclear power, for fear that this might lead to a conflict that would escalate all the way to nuclear war.

Nuclear weapons and cyber attacks don’t mix well.

Unfortunately, this means that the advantages of cyber operations become an important liability for nuclear deterrence when they are used for “left of launch” attacks on nuclear launch systems. By secretly penetrating another state’s launch system, you may undermine the stable system of beliefs that discourages an attack.

Consider what might happen in a tense standoff between two states that both have nuclear weapons, where one state has penetrated the other state’s launch system, so that it could stop a nuclear counter attack. The state that has penetrated the launch system knows that it has a military advantage. However, if  it reveals the advantage to the target state, the target state will be able to patch its system, destroying the advantage. The target state does not know that it is at a disadvantage, and it cannot be told by the attacker.

The result is that the two states have very different perceptions of the situation. The attacker thinks that there is an imbalance of power, where it has the advantage. The target thinks that there is a balance of power, where both states have nuclear weapons and can deter each other. This means that the first state will be less likely to back down, and might escalate conflict, secure in the knowledge that it can neutralize the other state if necessary. However, the target state may too behave in provocative ways that raise the stakes, since it mistakenly believes that at a certain point the other state will have to back down, for fear of nuclear war. Thus, this creates a situation where each side may be more willing to escalate the tense situation, making it more likely that one state will decide to move toward war.

That is from Eric Gartzke and Jon R. Lindsay, there is more interesting material at the link.

This 23-year-old Filipino-American should be starring in a Malcolm Gladwell column.  A few years ago, he was an up-and-coming aggressive, tactical chess prodigy, presumably lacking in the strategic niceties of the game at its highest levels.  I recall John Nye, my Filipino-American colleague (and chess player) coming into my office to discuss the astonishing fact that So had risen to number 9 in the world rankings.  I suggested that a bit of regression to the mean was in order, and So would not be returning to the top ten anytime soon.

Since that time, So has won four top tournaments in a row, besting Magnus Carlsen, and has had a 56-game non-losing streak, against very high caliber players, and recently he was selected best chess player of the last year.  Arguably he is the second best player in the world, and the one most likely to dethrone Carlsen from the world championship.

A turning point for So came in 2014 when he left university and moved to Minnetonka, Minnnesota to live with his adoptive parents, Lotis Key and Renato Kabigting, Key being a former Filipino movie star and now Vice President of the Minnesota Christian Writers Guild.  She serves as So’s manager and insisted he not check on-line NBA scores when doing his chess training.  Later, So turned away from the internet more fundamentally to focus on chess.

One Filipino international master remarked: “He cannot afford decent training given by well known GM-coaches and has to rely on his pure talent.”  Last month he brought on Vladimir Tukmakov as a coach, but he’s had less formal training than any top player in recent memory.

So hopes to learn how to drive a car, and he enjoys playing in Las Vegas: “I like Las Vegas,” So said, laughing. “People are usually drunk. Makes them easier to beat. Just keep drinking.”

So’s style now has evolved to the point where…he doesn’t seem to have a style.  He is renowned for his calm and he simply limits the number of mistakes.  At the top, top levels, a player without a real style is a player who is hard to train for and hard to beat.

So is religious, and he is considered mild-mannered and humble.  The story of Wesley So is not over.  Yet Wesley So is an American, and an American hero, and he has received virtually no mainstream media attention.

The basic post is too long, but some of it is interesting and here is the best part:

The pattern of Cost Disease seems to be related to things that inextricably require the unsubstitutable labour and attention not just of human beings but of human beings somehow comparable to the buyer. (Americans, for the US focus of most of this discussion.) Education not only requires teachers who are part of the same cultural milieu as their students, but it requires the attention of the students themselves, and attention is inherently expensive. As the only thing that can be expensive in the final Strong Heaven, attention predictably gets more expensive in a culture that moves more and more toward general post-scarcity. Health care similarly requires local human involvement.

That is from Ansuz, via Matthew Fairbank.  And here is Scott Alexander’s survey follow-up post.

In the United States, if I am trying to accelerate and enter the road from say a parking lot, I try to minimize the number of misleading movements my car might make.  I don’t “edge out” just for the heck of it, for fear this may spook the other drivers and cause them to suddenly switch lanes in a dangerous (for them) fashion.  Furthermore, I might misjudge and move the car out too far into the lane, leading to a collision.

In Lagos, it seems the practice is to announce your intentions with as many little forward “nudges” of your car as possible.  They seem to mean “I am thinking of going sometime soon.”

After enough such nudges, the oncoming cars either go far away into the left lane, or perhaps they stop for you altogether and let you go.  Or maybe they slow down a bit and you decide you can beat them and so you pull into the lane.

A higher discount rate (for entering the road) is one way to rationalize this behavior, but in a variety of other contexts I have noticed Nigerians who were not massively upset at being ever so slightly late.  So might there be some other explanation?

Maybe there is greater variability in rational assumptions about the other drivers.  You may not know how well their cars can brake, accelerate, and perhaps their lane-switching plans and propensities are harder to predict.  So by nudging your car out in successive bits, you may be “taking the temperature” of the other drivers on the road.  Keep in mind that they, too, may not have a good sense of how well your car can accelerate (furthermore some of the vehicles are tuks-tuks, not cars).  A willingness to make more nudges may be telling the other drivers that your engine is pretty good and your will is strong.

So they read your nudging pattern, and you draw inferences from their lane-switching and stopping responses.  Ex post (one hopes), everyone has a better sense of what the other cars and drivers are capable of doing.

Of course this is speculative.  The key point here is that greater variability in potential performance creates a case for sending more and smaller bits of the signal in advance.

In 1987, Trump made his goal of Russian collaboration on nuclear power explicit: The Soviet Union and the US should partner to form a nuclear superpower with the intention of intimidating other countries into dropping their own nuclear plans.

“Most of those [pre-nuclear] countries are in one form or another dominated by the US and the Soviet Union,” Trump told journalist Roy Rosenbaum. “Between those two nations you have the power to dominate any of those countries. So we should use our power of economic retaliation and they use their powers of retaliation, and between the two of us we will prevent the problem from happening. It would have been better having done something five years ago. But I believe even a country such as Pakistan would have to do something now. Five years from now they’ll laugh.”

Nuclear-related sanctions, from the two major powers, were to be applied to both Pakistan and France [sic].  Here is the full article, I cannot vouch for this account or any particular interpretation of it, but the hypothesis is new to me and so I present it to you as well.

Here is my 2005 tribute post, Thomas Schelling, Nobel Laureate.  Here are all the other MR posts about Schelling.  A great loss of course and for me he was a true role model and important advisor throughout my career.  Here is the University of Maryland notice.

The gambling market is somewhat saturated, so how can new customers be found?

One idea: skills-based games.

In Atlantic City, the Borgata added a basketball free-throw shooting contest. Other casinos are adding skill-based games to electronic slot machines — shooting, puzzles, less slot machine ding ding ding and more Angry Birds-style competition.

Maryland does not allow such games yet, but the state’s gaming agency says it is working on the issue.

That is from Michael Rosenwald, with most of the article covering D.C.’s foray into the casino genre.

Carlsen > Karjakin

by on November 30, 2016 at 10:04 pm in Games | Permalink

1. Karjakin played some of the best defensive chess ever, finding resources where there appeared to be none.

2. Carlsen had become a bit lazy, relying too much on his stamina advantage to beat opponents (yes I do understand that is an odd notion of lazy!).  Yet he had no real stamina advantage over Karjakin, who is of the same age and came to the match in very good physical shape.  So Carlsen simply could not grind him down, and it took Carlsen the entire match to realize that.

3. Karjakin made very few attempts to achieve demonstrable, sharp advantages.  That limited his total number of victories to one.

4. In the rapid tie-breaker — four consecutive games in the final day — Carlsen couldn’t try to win on stamina and simply showed he was the better player across many dimensions of the game.  Karjakin posed him no problems at all in these contests.

5. Karjakin played as Carlsen’s equal for the twelve regular time control games.  Yet I don’t think he will be back as a challenger.  His style is too “drab” (Kasparov’s description) to get through all of the risk-rewarding tournaments to reach the final championship match again.

6. Perhaps rapid chess is the future of chess as a spectator sport.  Four games in a row, each twenty-five minutes per player, plus increments.  It was thrilling, and I watched on the train.

7. Putin finally lost one this year, let’s hope this reverses the trend.

Here is the Chessbase account, here is the quite good NYT story.

Magnus Carlsen was heavily favored, but after six games with Karjakin — half the match — they are all draws.  The first noteworthy feature of play is that Karjakin has matched what is usually a big stamina advantage for Carlsen.  If Carlsen were playing against Anand again, and creating similar positions, he probably would be two points up, as he was pressing strong endgame advantages for hours in two of the games.  Yet Karjakin held firm and found the necessary defensive resources in both cases and victories fell from Carlsen’s grasp.  Overall we have seen few obvious mistakes in the play.

The second noteworthy feature of the first half of the match is that Carlsen has shown no inferiority of opening preparation, unlike what is usually the case; if anything Carlsen has had slightly better prep.  (And note that Karjakin has Putin and thus an army of seconds on his side; Karjakin has a connection to Crimea and used to play for Ukraine but now plays for Russia and basically endorsed the territory transfer, get the picture?)

I believe one should root for Carlsen.  Yet this is 2016, and I suppose anything could happen at this point…

Turkey continues a major crackdown against the Kurds.

This week at least 239 people drowned trying to cross the Mediterranean.

Huawei will begin marketing its smart phone in the United States this January.

Journalists will try to tell you that Carlsen vs. Karjakin will be close.  Other journalists will try to tell you that someone other than Golden State or Cleveland will win the next NBA title.  Other journalists will try to tell you…

“U.S. military hackers have penetrated Russia’s electric grid, telecommunications networks and the Kremlin’s command systems, making them vulnerable to attack by secret American cyber weapons should the U.S. deem it necessary”…link here.

TaterGrams: New Alberta company lets you mail personalized potatoes.

tatergram

Just keep Mexico, South Korea, and Estonia in mind, and I’m sure you’ll do the right thing.

No, I don’t see them voting down Brexit, any more than Republican Senators ever were going to endorse Hillary Clinton, even though many of them are rooting for her.  The more likely scenario out of Brexit is simply that Parliament stalls, demanding that Theresa May give them “the right Brexit.”  Of course there is no such thing, wrong Brexit is wrong Brexit, if only because EU-27 cannot agree on very much.  But with enough stalling, eventually another national election will be held and of course Brexit would be a major issue, probably the major issue.  That in essence would serve as a second referendum, and if anti-Brexit candidates did well enough, parliamentarians would have cover to go against the previous expression of the public will.

I give that path out of Brexit p = 0.2, with another p = 0.1 for “somehow Brexit just doesn’t happen.”

Here is commentary from Joshua Tucker.  And from Jolyom Maugham.

Addendum: The final word on Brexit rights may be held by…the European Court of Justice.

There is no grape stagnation

by on October 27, 2016 at 12:23 am in Games, Sports | Permalink

Setting the world record for using your mouth to catch a grape dropped from the greatest height: It was a dream years in the making, and all it took was a hot air balloon, walnut-sized fruits shipped specially from Georgia and a crew of Ph.D.-level engineers who gathered at a tiny Vermont airport before the sun rose on Monday morning.

The man with the plan was Brent Fraser, 35, who said he “just had a natural knack” for catching things in his mouth ever since his high school days in Barre, Vt., where buddies would chuck food toward him in the school parking lot.

The piece has some good sentences, such as:

Indeed, once things did get going, most of the few dozen attempts ended with a goggle-clad Fraser getting smacked in the face and chest by the large grapes — selected because they were easiest to see — that were traveling about 56 mph.

And:

“How much did they hurt?” one of the engineers, Tristan Ramey, asked at one point.

“So bad,” Fraser told her. “I felt like I was being punched in the face.”

He ended up catching one from 101 feet.  And finally:

Fraser, most of his face stained in purple grape juice, had to get to work to interview a prospective employee by 9 a.m.

Here is the full article, with video, via the excellent Mark Thorson.

My Twitter feed is mocking the policy behind this news report, but of course it makes perfect sense.  Here goes:

1. The United States wishes to have it be common knowledge that it can embarrass Putin.  But in fact maybe it can’t!  (At least not with a policy we are willing to bear the consequences of.)  So why not threaten that you can?  A truly secret strike probably would hurt him less than an embarrassment.  So start investing in the embarrassment now.

2. If the U.S. does do something cyber against Russia, it may wish to signal in advance that it won’t be truly severe, so as to limit retaliation and lower the probability of ongoing escalation.  Some public discussion can achieve this end.  Truly devastating blows are in fact usually delivered in secret.

3. There is a chance that the U.S. can’t/won’t do much if anything against Russia at all.  In that case third parties (Iran, China) may not know this for sure, and the announcement may have a slight deterrent value in their direction.

4. It may not be possible to understand the entire American strategy without knowing the private messages that are being sent to Putin at the same time.  For instance, the overall strategy may be “announce a coming mild retaliation and privately threaten a more severe action.”  Is that really so out of place?  Probably not.

In other words, “announced secrets” sometimes can make perfect sense.

According to Donner: “The whole point of the game [is] to prevent an artistic performance.” The former world champion Garry Kasparov makes the same point. “The highest art of the chess player,” he says, “lies in not allowing your opponent to show you what he can do.” Always the other player is there trying to wreck your masterpiece. Chess, Donner insists, is a struggle, a fight to the death. “When one of the two players has imposed his will on the other and can at last begin to be freely creative, the game is over. That is the moment when, among masters, the opponent resigns. That is why chess is not art. No, chess cannot be compared with anything. Many things can be compared with chess, but chess is only chess.”

That is Stephen Moss at The Guardian.  Along related lines, I very much enjoyed Daniel Gormally’s Insanity, Passion, and Addiction: A Year Inside the Chess World.  It’s one of my favorite books of the year so far, but it’s so miserable I can’t recommend it to anyone.  It’s a book about chess, and it doesn’t even focus on the great players.  It’s about the players who are good enough to make a living — ever so barely — but not do any better.  It serves up sentences such as:

Surely the money in chess is so bad that this can’t be all you do for a living?  But in fact in my experience, the majority of chess players rated over 2400 tend to just do chess.  If not playing, then something related to it, like coaching or DVDs.  That’s because we’re lazy, so making the monumental effort of a complete change in career is just too frightening a prospect.  So we stick with chess, even though the pay tends to be lousy, because most of our friends and contacts are chess players.  Our life is chess.  As a rough estimate, I would say there are as many 2600 players making less than £20,000 a year.

And:

Stability. I had this conversation with German number one Arkadij Naidisch at a blitz tournament in Scotland about a year ago. (there I go, name-dropping again.)  He suggested that a lot of people don’t achieve their goals because they just aren’t stable enough.  They’ll have a fantastic result somewhere, but then that’ll be let down by a terrible tournament somewhere else.

…The problem is it’s hard to break out of the habits of a lifetime.  Many times at home I’ve said to myself while sitting around depressed about my future and where my chess is going, “tomorrow will be different.  I’ll get up and study six-eight hours studying chess.”  But it never happens.

Overall biography and autobiography are far too specialized in the lives of the famous and successful.