Played with atoms and photons rather than dice and coins, quantum games
are contests whose outcomes are governed by the unusual logic of the
submicroscopic world. The basic token in a quantum game is a "qubit,"
a bit of data which is stored in an object such as an atomic nucleus.
While a classical coin can only be heads (data value 0) or tails (data
value 1), a qubit can effectively be both heads (0) and tails (1) at
the same time, since the nucleus can be in a combination or superposition
of spin-up (0) and spin-down (1). What's more, one can interlink or
"entangle" qubits held by separate players so that manipulating
one qubit strongly affects the others.
More than a diversion, playing quantum games can reveal new information-processing
tasks (possibly even certain types of financial transactions) that quantum
computers could perform more efficiently than classical computers. Towards
these ends, theorists have been taking traditional games, adapting them
for the quantum realm, and checking if new or better strategies emerge
for winning.
While past quantum games have focused on two players (Update 411),
Oxford researchers (Patrick Hayden, patrick.hayden@qubit.org) have now
identified multiplayer games in which the player's optimal strategy
differs from that of the classical version of the game. The researchers
discovered unique strategies in a three-player quantum version of the
Dilemma game, in which three partners engaged in a venture (such as
getting the best seats at a concert) each decide whether or not to betray
the others in efforts to maximize personal gain.
In the quantum version, the qubits are entangled, then each person
uses his qubit to choose between the following strategies: try for good
seat (0), settle for poor seat (1) or some superposition of the two.
Entanglement actually destroys the incentive for a player to contradict
and thereby betray his opponents and it removes the classical dilemma
entirely. Although quantum games are mostly played on paper at this
point, a Chinese group has just reported the experimental realization
of a quantum Prisoner's Dilemma (Los
Alamos preprint). (Benjamin
and Hayden, Physical Review A, September 2001.)