We consider multiarmed bandit problems where the expected reward is unimodal over partially ordered arms. In particular, the arms may belong to a continuous interval or correspond to vertices in a graph, where the graph structure represents similarity in rewards. The unimodality assumption has an important advantage: we can determine if a given arm is optimal by sampling the possible directions...

We consider the problem of decomposing a compactly supported distribution f : Rn → [0,∞) into a minimal number of unimodal components by means of some convex operation (e.g., sum or sup). The resulting “unimodal category” of f is a topological invariant of the distribution which shares a number of properties with the Lusternik-Schnirelmann category of a topological space. This work introduces t...

This module provides an overview of multimodal perception, including information Your nose might even be stimulated by the smell of burning rubber or gasoline. In other words, how does the perceptual system determine which unimodal between the two balls that then bounce off each other in opposite directions. Principles and heuristics for designing minimalist instruction. H Van der Multimodal ve...

A finite sequence of real numbers {d0, d1, · · · , dm} is said to be unimodal if there exists an index 0 ≤ j ≤ m such that d0 ≤ d1 ≤ · · · ≤ dj and dj ≥ dj+1 ≥ · · · ≥ dm. A polynomial is said to be unimodal if its sequence of coefficients is unimodal. The sequence {d0, d1, · · · , dm} with dj ≥ 0 is said to be logarithmically concave (or log concave for short) if dj+1dj−1 ≤ dj for 1 ≤ j ≤ m − ...