Konstantin Rybnikov

Criteria for Balance in Abelian Gain Graphs, with Applications to Piecewise-Linear Geometry

AbstractConsider a gain graph with abelian gain group having no odd torsion. If there is a basis of the graph’s binary cycle space, each of whose members can be lifted to a closed walk whose gain is the identity, then the gain graph is balanced, provided that the graph is finite or the group has no non-trivial infinitely 2-divisible elements. We apply this theorem to deduce a result on the projective geometry of piecewise-linear realizations of cell-decompositions of manifolds.

Confident Predictability: Identifying reliable gene expression patterns for individualized tumor classification using a local minimax kernel algorithm

BackgroundMolecular classification of tumors can be achieved by global gene expression profiling. Most machine learning classification algorithms furnish global error rates for the entire population. A few algorithms provide an estimate of probability of malignancy for each queried patient but the degree of accuracy of these estimates is unknown. On the other hand local minimax learning provides such probability estimates with best finite sample bounds on expected mean squared error on an individual basis for each queried patient. This allows a significant percentage of the patients to be identified as confidently predictable, a condition that ensures that the machine learning algorithm possesses an error rate below the tolerable level when applied to the confidently predictable patients.ResultsWe devise a new learning method that implements: (i) feature selection using the k-TSP algorithm and (ii) classifier construction by local minimax kernel learning. We test our method on three publicly available gene expression datasets and achieve significantly lower error rate for a substantial identifiable subset of patients. Our final classifiers are simple to interpret and they can make prediction on an individual basis with an individualized confidence level.ConclusionsPatients that were predicted confidently by the classifiers as cancer can receive immediate and appropriate treatment whilst patients that were predicted confidently as healthy will be spared from unnecessary treatment. We believe that our method can be a useful tool to translate the gene expression signatures into clinical practice for personalized medicine.

A New Algorithm in Geometry of Numbers

A lattice Delaunay polytope P is called perfect if its Delaunay sphere is the only ellipsoid circumscribed about P. We present a new algorithm for finding perfect Delaunay polytopes. Our method overcomes the major shortcomings of the previously used method [Du05]. We have implemented and used our algorithm for finding perfect Delaunay polytopes in dimensions 6, 7, 8. Our findings lead to a new conjecture that sheds light on the structure of lattice Delaunay tilings.