Predicting the Labelling of a Graph via Minimum $p$-Seminorm Interpolation
نویسندگان
چکیده
We study the problem of predicting the labelling of a graph. The graph is given and a trial sequence of (vertex,label) pairs is then incrementally revealed to the learner. On each trial a vertex is queried and the learner predicts a boolean label. The true label is then returned. The learner’s goal is to minimise mistaken predictions. We propose minimum p-seminorm interpolation to solve this problem. To this end we give a p-seminorm on the space of graph labellings. Thus on every trial we predict using the labelling which minimises the p-seminorm and is also consistent with the revealed (vertex, label) pairs. When p = 2 this is the harmonic energy minimisation procedure of [21], also called (Laplacian) interpolated regularisation in [1]. In the limit as p → 1 this is equivalent to predicting with a label-consistent mincut. We give mistake bounds relative to a label-consistent mincut and a resistive cover of the graph. We say an edge is cut with respect to a labelling if the connected vertices have disagreeing labels. We find that minimising the p-seminorm with p = 1 + where → 0 as the graph diameter D → ∞ gives a bound of O(Φ logD) versus a bound of O(ΦD) when p = 2 where Φ is the number of cut edges.
منابع مشابه
On the bandwidth of Mobius graphs
Bandwidth labelling is a well known research area in graph theory. We provide a new proof that the bandwidth of Mobius ladder is 4, if it is not a $K_{4}$, and investigate the bandwidth of a wider class of Mobius graphs of even strips.
متن کاملOn the Error of Linear Interpolation and the Orientation, Aspect Ratio, and Internal Angles of a Triangle
In this paper, we attempt to reveal the precise relation between the error of linear interpolation on a general triangle and the geometric characters of the triangle. Taking the model problem of interpolating quadratic functions, we derive two exact formulas for the H1-seminorm and L2-norm of the interpolation error in terms of the area, aspect ratio, orientation, and internal angles of the tri...
متن کاملOptimal Mesh for P1 Interpolation in H1 Seminorm
In this paper we present one approach to build optimal meshes for P1 interpolation. Considering classical geometric error estimates based on the Hessian matrix of a solution, we show it is possible to generate optimal meshes in H 1 semi-norm via a simple minimization procedure.
متن کاملSobolev seminorm of quadratic functions with applications to derivative-free optimization
This paper studies the H1 Sobolev seminorm of quadratic functions. The research is motivated by the least-norm interpolation that is widely used in derivative-free optimization. We express the H1 seminorm of a quadratic function explicitly in terms of the Hessian and the gradient when the underlying domain is a ball. The seminorm gives new insights into least-norm interpolation. It clarifies th...
متن کامل