Cai and Furst introduced the notion of bottleneck Turing machines and showed that the languages recognized by width-5 bottleneck Turing machines are exactly those in PSPACE. Computational power of bottleneck Turing machines with width fewer than 5 is investigated. It is shown that width-2 bottleneck Turing machines capture polynomial-time many-one closure of nearly near-testable sets. For langu...

We present a variational approximation to the information bottleneck of Tishby et al. (1999). This variational approach allows us to parameterize the information bottleneck model using a neural network and leverage the reparameterization trick for efficient training. We call this method “Deep Variational Information Bottleneck”, or Deep VIB. We show that models trained with the VIB objective ou...

Let R be a set of n red segments and B a set of n blue segments, we wish to find the minimum value d∗, such that there exists a perfect matching between R and B with bottleneck d∗, i.e., the maximum distance between a matched red-blue pair is d∗. We first solve the corresponding decision problem: Given R, B and a distance d > 0, find a maximum matching between R and B with bottleneck at most d....

To improve speech recognition performance, a combination between TANDEM and bottleneck Deep Neural Networks (DNN) is investigated. In particular, exploiting a feature combination performed by means of a multi-stream hierarchical processing, we show a performance improvement by combining the same input features processed by different neural networks. The experiments are based on the spontaneous ...

The bottleneck cycle refers to the iterative process of finding a performance bottleneck, eliminating it, and finding the next one. The goal is that eventually the bottleneck falls outside of your application’s performance requirements. This paper will present a case study to illustrate how using the principles of application performance testing enabled us to traverse our way through the bottle...

Algorithms for the constrained bottleneck matroid base problem and the constrained bottleneck independent set problem are presented. The algorithms have improved complexities upon the best known general algorithm. Sufficient conditions are obtained when the proposed algorithms have linear complexity and are therefore the best possible ones. Several specific classes of matroidal problems are con...

We present a reformulation of the information bottleneck (IB) problem in terms of copula, using the equivalence between mutual information and negative copula entropy. Focusing on the Gaussian copula we extend the analytical IB solution available for the multivariate Gaussian case to distributions with a Gaussian dependence structure but arbitrary marginal densities, also called meta-Gaussian d...

We consider the online bottleneck matching problem, where k serververtices lie in a metric space and k request-vertices that arrive over time each must immediately be permanently assigned to a server-vertex. The goal is to minimize the maximum distance between any request and its server. Because no algorithm can have a competitive ratio better thanO(k) for this problem, we use resource augmenta...

Urban road congestions change both temporally and spatially. They are essentially caused by network bottlenecks. Therefore, understanding bottleneck dynamics is critical in the goal of reasonably allocating transportation resources. In general, a typical bottleneck experiences the stages of formation, propagation and dispersion. In order to understand the three stages of a bottle neck and how t...