• Publications
  • Influence
Finite-time Analysis of the Multiarmed Bandit Problem
This work shows that the optimal logarithmic regret is also achievable uniformly over time, with simple and efficient policies, and for all reward distributions with bounded support. Expand
The complexity of computing the MCD-estimator
Here a polynomial time algorithm for MCD for fixed dimension of the data is presented and it is shown that computing the MCD-estimator is NP-hard if the dimension varies. Expand
Sequential and Parallel Algorithms for Finding a Maximum Convex Polygon
  • P. Fischer
  • Computer Science, Mathematics
  • Comput. Geom.
  • 1 February 1997
It is shown how to find a maximum convex polygon which contains a given point in time O(n3logn), and how to use other measures fulfilling a certain additive property, however, this may affect the running time. Expand
On learning ring-sum-expansions
It is proved that 2-term is learnable by a conjunction of a 2-CNF and a 1-DNF and that k-RSE, the class of ring-sum-expansions containing only monomials of length at most k, can be learned from positive (negative) examples alone. Expand
On the Cut-off Point for Combinatorial Group Testing
It has been conjectured that the cut-off point of combinatorial group testing is equal to x ∗ = 1 3 and the strategy of testing n − 1 objects individually minimizes the worst case number of queries iff k n ⩾ α ∗ and k. Expand
Comparison between various regression depth methods and the support vector machine to approximate the minimum number of missclassifications
Two new approaches to approximating the minimum number of misclassifications achievable with affine hyperplanes are introduced, modifications of the regression depth method proposed by Rousseeuw and Hubert (1999) for linear regression models. Expand
Finding Maximum Convex Polygons
  • P. Fischer
  • Mathematics, Computer Science
  • FCT
  • 23 August 1993
This paper considers the situation where one is given a finite set of n points in the plane each of which is labeled either “positive” or “negative” and shows that the problem to find a bounded convex polygon of maximum area can be solved in time O(n4 log n). Expand
Edge-matching Problems with Rotations
It is shown that some problems have polynomial time algorithms while others are NP-complete, and that many commonly considered puzzles can be emulated by simple puzzles with quadratic pieces, so that one can restrict oneself to investigating those. Expand
Sample-efficient strategies for learning in the presence of noise
This paper presents a generic algorithm using randomized hypotheses that can tolerate noise rates slightly larger than ε/(1 + ε) while using samples of size <italic>d</italic*/ε as in the noise-free case and shows upper and lower bounds of this order. Expand