Looks great! Next step is to do like AlphaGo / AlphaStar and use the MCTS data to train a neural network to act as the value function.
7 karma · joined June 13, 2014
The joke is based on a logical paradox known as the *barber paradox*¹. The barber paradox is a situation where a statement leads to a contradiction when trying to determine its truth value. In this case, the statement is "none of you want tequila shots?".
If none of the logicians want tequila shots, then they should all answer "yes" to confirm the statement. But if any of them answers "yes", then it means that at least one of them wants tequila shots, which contradicts the statement. So they cannot answer "yes".
If any of the logicians want tequila shots, then they should answer "no" to deny the statement. But if all of them answer "no", then it means that none of them want tequila shots, which confirms the statement. So they cannot answer "no".
The first two logicians avoid this contradiction by answering "I don't know". They are not saying whether they want tequila shots or not, but only whether they know if none of them want tequila shots or not. By answering "I don't know", they are implying that their answer depends on what the other logicians say.
The third logician breaks this impasse by answering "No thank you". He is not answering the question directly, but politely declining the offer of tequila shots. By doing so, he reveals that he does not want tequila shots, and also implies that he knows that none of them want tequila shots. This makes his answer consistent with both possibilities: if none of them want tequila shots, he is confirming the statement; if any of them want tequila shots, he is denying the statement.
The joke is funny because it plays with the different levels of logic and language involved in answering a simple question. It also shows how logicians can use their skills to avoid contradictions and confusions.
Java 8 Lambdas provide such an elegant syntax for anonymous classes... When lambdas were first introduced in Java , a lot of people were underwhelmed by them. It is hard to evaluate at first the impact they can have on a well designed API.