Learn More
We show that a polynomial-time rescaling algorithm, in [5], for finding feasible solutions to linear programs, has flaws in its theoretical framework. With new estimates for the algorithm, we show that it indeed terminates in polynomial time. Also, we present a modification of the algorithm, where we suggest a data dependent way of choosing a rescaling(More)
  • Olof Barr
  • 2005
We construct a modied perceptron algorithm that is deter-ministic, polynomial and also as fast as previous known algorithms. The algorithm runs in time O(mn 3 log n log(1/ρ)), where m is the number of examples, n the number of dimensions and ρ is approximately the size of the margin. We also construct a non-deterministic modied perceptron algorithm running(More)
  • 1