The unbeatability of a consensus protocol, introduced by Halpern, Moses and Waarts in [14],is a stronger notion of optimality than the accepted notion of early stopping protocols. Usinga novel knowledge-based analysis, this paper derives the first practical unbeatable consensusprotocols in the literature, for the standard synchronous message-passing model with crashfailures. The...

Random Early Detection (RED) can stabilize the queue within a given target range and thus achieve high throughput in the routers. However, the average queue size is quite sensitive to network scenarios and it is difficult to adapt RED parameters to changing network traffic. In this paper we use a previously developed dynamic model of TCP behavior and linear feedback model of TCP/RED to analyze ...

The first passage probability (FPP), of trafficked intracellular particles reaching a displacement L, in a given time t or inverse velocity S = t/L, can be calculated robustly from measured particle tracks. The FPP gives a measure of particle movement in which different types of motion, e.g. diffusion, ballistic motion, and transient run-rest motion, can readily be distinguished in a single gra...

A TCP/RED (Transmission Control Protocol/Random Early Detection) system with multiplebottleneck links could be unstable even if its system parameters are set the same as those in a stable single-bottleneck system [D. Bauso, L. Giarre, G. Neglia, Active queue management stability in multiple bottleneck networks, IEEE ICC’04, vol. 4, June 2004, pp. 2267– 2271]. In this paper, we study the stabili...

This paper presents a summary of our work on developing an architecture for transporting real-time traffic (MPEG4 video in this paper) in IP networks that provide service differentiation. We target our architecture at Assured Forwarding (AF) style services. This architecture assumes loss differentiation in the network and the network’s ability to provide ECN messages to the sender. We did not c...

Early stopping is a well known approach to reduce the time complexity for performing training and model selection of large scale learning machines. On the other hand, memory/space (rather than time) complexity is the main constraint in many applications, and randomized subsampling techniques have been proposed to tackle this issue. In this paper we ask whether early stopping and subsampling ide...

Distributed communication systems have gained a substantial importance over the past years with a large set of examples of systems that are present in our everyday life. The heterogeneity of applications and application domains speaks for the complexity of such systems and the challenges that developers are faced with. The focus of this dissertation is on the development of applications for dis...

Earthquake Alarm Systems, or ElarmS, is a network-based earthquake early warning (EEW) system developed in California. This dissertation concerns the transition of ElarmS from research prototype to production-grade code. We test ElarmS' performance for large earthquakes with an extensive dataset of large events from Japan. Using the Japanese dataset and results, we develop a statistical error m...

A theory of early stopping as applied to linear models is presented. The backpropagation learning algorithm is modeled as gradient descent in continuous time. Given a training set and a validation set, all weight vectors found by early stopping must lie on a certain quadric surface, usually an ellipsoid. Given a training set and a candidate early stopping weight vector, all validation sets have...

In active queue management (AQM), core routers signal transmission control protocol (TCP) sources with the objective of managing queue utilization and delay. It is essentially a feedback control problem. Based on a recently developed dynamic model of TCPs congestion-avoidance mode, this paper does three things. First, it relates key network parameters such as the number of TCP sessions, link ca...