Balanced allocations
Yossi Azar, Andrei Z. Broder, et al.
SIAM Journal on Computing
We introduce backoff processes, an idealized stochastic model of browsing on the World Wide Web, which incorporates both hyperlink traversals and use of the "back button." With some probability the next state is generated by a distribution over out-edges from the current state, as in a traditional Markov chain. With the remaining probability, however, the next state is generated by clicking on the back button and returning to the state from which the current state was entered by a "forward step." Repeated clicks on the back button require access to increasingly distant history. We show that this process has fascinating similarities to and differences from Markov chains. In particular, we prove that, like Markov chains, back-off processes always have a limit distribution, and we give algorithms to compute this distribution. Unlike Markov chains, the limit distribution may depend on the start state.
Yossi Azar, Andrei Z. Broder, et al.
SIAM Journal on Computing
Ronald Fagin, Phokion G. Kolaitis, et al.
SIGMOD/PODS/ 2009
Moses Charikar, Jon Kleinberg, et al.
SIAM Journal on Discrete Mathematics
Alok Aggarwal, Jon Kleinberg, et al.
SIAM Journal on Computing