Bottleneck Routing Games in Communication Networks

We consider routing games where the performance of each user is dictated by the worst (bottleneck) element it employs. We are given a network, finitely many (selfish) users, each associated with a positive flow demand, and a load-dependent performance function for each network element; the social (i.e., system) objective is to optimize the performance of the worst element in the network (i.e., the network bottleneck). Although we show that such "bottleneck" routing games appear in a variety of practical scenarios, they have not been considered yet. Accordingly, we study their properties, considering two routing scenarios, namely when a user can split its traffic over more than one path (splittable bottleneck game) and when it cannot (unsplittable bottleneck game). First, we prove that, for both splittable and unsplittable bottleneck games, there is a (not necessarily unique) Nash equilibrium. Then, we consider the rate of convergence to a Nash equilibrium in each game. Finally, we investigate the efficiency of the Nash equilibria in both games with respect to the social optimum; specifically, while for both games we show that the price of anarchy is unbounded, we identify for each game conditions under which Nash equilibria are socially optimal.     

Multipath Routing Algorithms for Congestion Minimization

Unlike traditional routing schemes that route all traffic along a single path, multipath routing strategies split the traffic among several paths in order to ease congestion. It has been widely recognized that multipath routing can be fundamentally more efficient than the traditional approach of routing along single paths. Yet, in contrast to the single-path routing approach, most studies in the context of multipath routing focused on heuristic methods. We demonstrate the significant advantage of optimal (or near optimal) solutions. Hence, we investigate multipath routing adopting a rigorous (theoretical) approach. We formalize problems that incorporate two major requirements of multipath routing. Then, we establish the intractability of these problems in terms of computational complexity. Finally, we establish efficient solutions with proven performance guarantees     

Pregnancy following a single lung transplant

Successful pregnancy in a single lung transplant recipient has not been reported previously. The long term effect of pregnancy on graft function and management of deteriorating pulmonary function is not defined. This case describes the management, outcome, and problems encountered when a single lung transplant recipient developed a progressive deterioration in pulmonary function during pregnancy, attributed to accelerated obliterative bronchiolitis.     

The Emerging Political Perception of Organizations: Two Paradigms

This article examines the premise that increasing concentration of power in large, multinational organizations will eventually necessitate the political control of such organizations to ensure the interests and well-being, however defined, of their host countries. This view assumes that centralized control is necessary to prevent abuses of power by private organizations who are not necessarily concerned with the public good. Another view is the anarchist paradigm, where it is assumed that people and groups, left to sort things out among themselves, will eventually produce outcomes beneficial for the whole. Aligned with this approach is the transformational or emerging paradigm. I will examine the assumptions that underlie each paradigm and leave the reader to decide their relative validity.     

In Situ Liquid-Cell TEM Observation of Multiphase Classical and Nonclassical Nucleation of Calcium Oxalate

Calcium oxalate (CaOx) is the major phase in kidney stones and the primary calcium storage medium in plants. CaOx can form crystals with different lattice types, water contents, and crystal structures. However, the conditions and mechanisms leading to nucleation of particular CaOx crystals are unclear. Here, liquid-cell transmission electron microscopy and atomistic molecular dynamics simulations are used to study in situ CaOx nucleation at different conditions. The observations reveal that rhombohedral CaOx monohydrate (COM) can nucleate via a classical pathway, while square COM can nucleate via a non-classical multiphase pathway. Citrate, a kidney stone inhibitor, increases the solubility of calcium by forming calcium-citrate complexes and blocks oxalate ions from approaching calcium. The presence of multiple hydrated ionic species draws additional water molecules into nucleating CaOx dihydrate crystals. These findings reveal that by controlling the nucleation pathways one can determine the macroscale crystal structure, hydration state, and morphology of CaOx.     

Understanding Unified Messaging

The proliferation of messaging systems presents users with a challenging task: how to effectively use and manage many separate messaging systems. Unified messaging provides simplified management of different messaging systems through a single, unified interface.     

A combined ion probe/spark source analysis system

Spark source mass spectrometry and secondary ion mass spectrometry have complementary advantages in the elemental analysis of solids. The AEI IM20 ion probe and MS702R mass spectrometer form a composite instrument incorporating both techniques; the design and different operating modes are described. A primary ion beam diameter as low as 2 μm can be obtained under favourable circumstances; the minimum detectable concentration of a typical element is derived as a function of this diameter. The mass resolution is greater than 5000 (10% valley definition) with electrical detection and 10,000 (50% density definition) with photographic detection; this resolution is much higher than that obtained with other secondary ion mass spectrometers. Two typical applications illustrate that such high mass resolution is a necessary prerequisite to quantitative interpretation in secondary ion mass spectrometry: precise isotopic analysis of titanium is demonstrated and the concentrations of impurity transition elements derived from ion probe and spark source analyses of a geological olivine are compared.     

Loss aware post-training quantization

Machine Learning - Neural network quantization enables the deployment of large models on resource-constrained devices. Current post-training quantization methods fall short in terms of accuracy for...