Big Red Bits

Today in a dinner with Thanh, Hu and Joel I heard about a paradox I haven’t heard so far. Probability is full of cute problems that challenge our understanding of the basic concepts. The most famous of them is the Monty Hall Problem, which asks:

You are on a TV game show and there are doors – one of them contains a prize, say a car and the other two door contain things you don’t care about, say goats. You choose a door. Then the TV host, who knows where the prize is, opens one door you haven’t chosen and that he knows has a goat. Then he asks if you want to stick to the door you have chosen or if you want to change to the other door. What should you do?

Probably you’ve already came across this question in some moment of your life and the answer is that changing doors would double your probability of getting the price. There are several ways of convincing your intuitions:

• Do the math: when you chose the door, there were three options so the prize is in the door you chose with probability and in the other door with probability (note that the presenter can always open some door with a goat, so conditioning on that event doesn’t give you any new information).
• Do the actual experiment (computationally) as done here. One can always ask a friend to help, get some goats and perform the actual experiment.
• To convince yourself that “it doesn’t matter” is not correct, think doors. You choose one and the TV host open of them and asks if you want to change or stick with your first choice. Wouldn’t you change?

I’ve seen TV shows where this happened and I acknowledge that other things may be involved: there might be behavioral and psychologic issues associated with the Monty Hall problem – and possibly those would interest Dan Ariely, whose book I began reading today – and looks quite fun. But the problem they told me about today in dinner was another: the envelope problem:

There are two envelopes and you are told that in one of them there is twice the amount that there is in the other. You choose one of the envelopes at random and open it: it contains bucks. Now, you don’t know if the other envelope has bucks or bucks. Then someone asks you if you wanted to pay bucks and change to the other envelope. Should you change?

Now, consider two different solutions to this problem: the first is fallacious and the second is correct:

1. If I don’t change, I get bucks, if I change I pay a penalty of and I get either or with equal probability, so my expected prize if I change is , so I should change.
2. I know there is one envelope with and one with , then my expected prize if I don’t change is . If I change, my expected prize is , so I should not change.

The fallacy in the first argument is perceiving a probability distribution where there is no one. Either the other envelope contains bucks or it contains bucks – we just don’t know, but there is no probability distribution there – it is a deterministic choice by the game designer. Most of those paradoxes are a result of either an ill-defined probability space, as Bertrand’s Paradox or a wrong comprehension of the probability space, as in Monty Hall or in several paradoxes exploring the same idea as: Three Prisioners, Sleeping Beauty, Boy or Girl Paradox, …

There was very recently a thrilling discussion about a variant on the envelope paradox in the xkcd blag – which is the blog accompaning that amazing webcomic. There was a recent blog post with a very intriguing problem. A better idea is to go there and read the discussion, but if you are not doing so, let me summarize it here. The problem is:

There are two envelopes containing each of them a distinct real number. You pick one envelope at random, open it and see the number, then you are asked to guess if the number in the other envelope is larger or smaller then the previous one. Can you guess correctly with more than probability?

A related problem is: given that you are playing the envelope game and there are number and (with ). You pick one envelope at random and then you are able to look at the content of the first envelope you open and then decide to switch or not. Is there a strategy that gives you expected earnings greater than ?

The very unexpected answers is yes !!! The strategy that Randall presents in the blog and there is a link to the source here is: let be a random variable on such that for each we have , for example, the normal distribution or the logistic distribution.

Sample then open the envelope and find a number now, if say the other number is lower and if say the other number is higher. You get it right with probability

which is impressive. If you follow your guess, your expected earning is:

The xkcd pointed to this cool archive of puzzles and riddles. I was also told that the xkcd puzzle forum is also a source of excellent puzzles, as this:

You are the most eligible bachelor in the kingdom, and as such the King has invited you to his castle so that you may choose one of his three daughters to marry. The eldest princess is honest and always tells the truth. The youngest princess is dishonest and always lies. The middle princess is mischievous and tells the truth sometimes and lies the rest of the time. As you will be forever married to one of the princesses, you want to marry the eldest (truth-teller) or the youngest (liar) because at least you know where you stand with them. The problem is that you cannot tell which sister is which just by their appearance, and the King will only grant you ONE yes or no question which you may only address to ONE of the sisters. What yes or no question can you ask which will ensure you do not marry the middle sister?

copied from here.

In my last post about hats, I told I’ll soon post another version with some more problems, which I ended up not doing and would talk a bit more about those kind of problems. I ended up not doing, but here are a few nice problems:

Those people are again a room, each with a hat which is either black or white (picked with probability at random) and they can see the color of the other people’s hats but they can’t see their own color. They write in a piece of paper either “BLACK” or “WHITE”. The whole team wins if all of them get their colors right. The whole team loses, if at least one writes the wrong color. Before entering the room and getting the hats, they can strategyze. What is a strategy that makes them win with probability?

If they all choose their colors at random, the probability of winning is very small: . So we should try to correlate them somehow. The solution is again related with error correcting codes. We can think of the hats as a string of bits. How to correct one bit if it is lost? The simple engineering solution is to add a parity check. We append to the string a bit . So, if bit is lost, we know it is . We can use this idea to solve the puzzle above: if hats are places with probability, the parity check will be with probability and with probability . They can decide before hand that everyone will use and with probability they are right and everyone gets his hat color right. Now, let’s extend this problem in some ways:

The same problem, but there are hat colors, they are choosen independently with probability and they win if everyone gets his color right. Find a strategy that wins with probability .

There are again hat colors, they are choosen independently with probability and they win if at least a fraction () of the people guesses the right color. Find a strategy that wins with probability .

Again to the problem where we just have BLACK and WHITE colors, they are chosen with probability and everyone needs to find the right color to win, can you prove that is the best one can do? And what about the two other problems above?

The first two use variations of the parity check idea in the solution. For the second case, given any strategy of the players, for each string they have probability . Therefore the total probability of winning is . Let , i.e., the same input but with the bit flipped. Notice that the answer of player is the same (or at least has the same probabilities) in both and , since he can’t distinguish between and . Therefore, . So,

. This way, no strategy can have more than probability of winning.

Another variation of it:

Suppose now we have two colors BLACK and WHITE and the hats are drawn from one distribution , i.e., we have a probability distribution over and we draw the colors from that distribution. Notice that now the hats are not uncorrelated. How to win again with probability (to win, everyone needs the right answer).

I like a lot those hat problems. A friend of mine just pointed out to me that there is a very nice paper by Bobby Kleinberg generalizing several aspects of hat problems, for example, when players have limited visibility of other players hats.

I began being interested by this sort of problem after reading the Derandomization of Auctions paper. Hat guessing games are not just a good model for error correcting codes, but they are also a good model for truthful auctions. Consider an auction with a set single parameter agents, i.e., an auction where each player gives one bid indicating how much he is willing to pay to win. We have a set of constraints: of all feasible allocations. Based on the bids we choose an allocation and we charge payments to the bidders. An example of a problem like this is the Digital Goods Auction, where .

In this blog post, I discussed the concept of truthful auction. If an auction is randomized, an universal truthful auction is an auction that is truthful even if all the random bits in the mechanism are revealed to the bidders. Consider the Digital Goods Auction. We can characterize universal truthful digital goods auction as bid-independent auctions. A bid-independent auction is given by function , which associated for each a random variable . In that auction, we offer the service to player at price . If we allocate to and charge him . Otherwise, we don’t allocate and we charge nothing.

It is not hard to see that all universal truthful mechanisms are like that: if is the probability that player gets the item bidding let be an uniform random variable on and define . Notice that here , but we are inverting with respect to . It is a simple exercise to prove that.

With this characterization, universal truthful auctions suddenly look very much like hat guessing games: we need to design a function that looks at everyone else’s bid but not on our own and in some sense, “guesses” what we probably have and with that calculated the price we offer. It would be great to be able to design a function that returns . That is unfortunately impossible. But how to approximate nicely? Some papers, like the Derandomization of Auctions and Competitiveness via Consensus use this idea.

When I was a child someone told me the following problem:

A king promised to marry his daughter to the most intelligent man. Three princes came to claim her hand and he tryed the following logic experiment with them: The princes are gathered into a room and seated in a line, one behind the other, and are shown 2 black hats and 3 white hats. They are blindfolded, and 1 hat is placed on each of their heads, with the remaining hats hidden in a different room. The first one to deduce his hat color will marry the princess. If some prince claims his hat color incorrectly he dies.

The prince who is seated behind removes his blindfold, sees the two hats in front of him and says nothing. Then the prince in the middle removes his blindfold after that and he can see the hat of the prince in front of him. He also says nothing. Noticing the other princes said nothing, the prince seated in the first whole, without even removing his blindfold, gives the correct answer? The question is: what is the color he said?

This is a simple logical puzzle: we just write all the possibilities and start ruling them out given that the other princes didn’t answer and in the end we can find the color of his hat. I remember that this puzzle surprised me a lot as a kid. A found it extremly cool by then, what made me want to read books about logic problems. After that, I had a lot of fun reading the books by Raymond Smullyan. I usually would read the problems, think something like: there can’t ba a solution to that. Then go to school with the problem in mind and spend the day thinking about that. Here is a problem I liked a lot:

There is one prisoner and there are two doors: each has one guardian. One door leads to an exit and one door leads to death. The prisioner can choose one door to open. One guardian speaks only the truth and one guardian always lies. But you don’t know which door is which, which guardian is which and who guards each door. You are allowed to choose one guardian and make him one Yes/No question, and then you need to choose a door. What is the right question to ask?

But my goal is not to talk about logic puzzles, but about Hat problems. There are a lot of variations of the problems above: in all of them a person is allowed to see the other hats but not his own hat and we need to “guess” which is the color of our hat. If we think carefully, we will see that this is a very general kind of problem in computer science: (i) the whole goal of learning theory is to predict one thing from a lot of other things you observe; (ii) in error correcting code, we should guess one bit from all the others, or from some subset of the others; (iii) in universal truthful mechanisms, we need to make a price offer to one player that just depends on all other players bids. I’ll come back to this example in a later post, since it is what made me interested in those kinds of problems, but for now, let’s look at one puzzle I was told about by David Malec at EC’09:

There are black and white hats and people: for each of them we choose one color independently at random with probability . Now, they can look at each others hats but not at their own hat. Then they need to write in a piece of paper either “PASS” or one color. If all pass or if someone has a wrong color, the whole team loses (this is a team game) and if at lest one person gets the color right and no one gets wrong, the whole team wins. Create a strategy for the team to win with probability.

To win with probability is easy: one person will always write “BLACK” and the others “PASS”. A better strategy is the following: if one person sees two hats of equal color, he writes the opposite color, otherwise, he passes. It is easy to see the team wins except in the case where all hats are the same color, what happens with probability. We would like to extend this to a more general setting:

There are black and white hats and people: for each of them we choose one color independently at random with probability . Now, they can look at each others hats but not at their own hat. Then they need to write in a piece of paper either “PASS” or one color. If all pass or if someone has a wrong color, the whole team loses (this is a team game) and if at lest one person gets the color right and no one gets wrong, the whole team wins. Create a strategy for the team to win with probability.

It is a tricky question on how to extend the above solution in that case. A detailed solution can be found here. The idea is quite ingenious, so I’ll sketch here. It envolves Error Correcting Code, in that case, the Hamming Code. Let with sum and product modulo . Let be the non-zero vector of and the following linear map:

Let be the kernel of that application. Then, it is not hard to see that is a partition of and also that because of that fact, for each either or exists a unique s.t. . This gives an algorithm for just one player to guess his correct color. Let be the color vector of the hats. Player sees this vector as:

which can be or . The strategy is: if either one of those vector is in , write the color corresponding to the other vector. If both are out of , pass. The team wins iff , what happens with probability. Is is an easy and fun exercise to prove those facts. Or you can refer to the solution I just wrote.

Now, we can complicate it a bit more: we can add other colors and other distributions. But I wanted to move to a different problem: the paper Derandomization of Auctions showed me an impressive thing: we can use coding theory to derandomize algorithms. To illustrate their ideas, they propose the following problem:

Color guessing problem: There are people wearing hats of different colors. If each person can see everyone else’s hats but not his or her own. Each person needs to guess the color of his own hat. We want a deterministic guessing algorithm that fraction of each color class is guessed correctly.

The problem is very easy if we have a source of random bits. Each person guesses some color at random. It seems very complicated to do that without random bits. Surprisingly, we will solve that using a flow computation:

Let be an array of colors the array with color removed. Consider the following flow network: nodes and (source and sink), nodes for each . There are such nodes. Consider also nodes in the form where is a color () and is a color vector. There are such nodes.

We have edges from to for all nodes of that kind. And we have edges from to . Now, if , i.e., if completed in the -th coordinate with generates , then add an edge from to .

Consider the following flow: add unit of flow from to and from split that flow in pieces of size and send each to for . Now, each node receives flow, where is the number of occurencies of in . Send all that flow to .

We can think of that flow as the guessing procedure. When we see we choose the guess independently at random and this way, each receives in expectation guesses . Notice that an integral flow in that graph represents a deterministic guessing procedure: so all we need is an integral flow so that the flow from to is . The flow received is from nodes of the type: and that means that bidder in , looking at the other hats will correctly choose , times.

Now, define the capacities this way: for all edges from to and from to have capacity and from to capacity . There is an integral flow that saturates all edges from to , because of the fractional flow showed. So, the solution gives us a deterministic decision procedure.

In the next blog post, I’ll try to show the result in the Derandomization of Auctions that relates that to competitive auctions.