/pdf/scalable-molecular-dynamics-with-namd-3j7xyc70g4.pdf

Scalable Molecular Dynamics with NAMD

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

The mapping of large-scale human movements is important for urban planning, traffic forecasting and epidemic prevention. Work in animals had suggested that their foraging might be explained in terms of a random walk, a mathematical rendition of a series of random steps, or a Levy flight, a random walk punctuated by occasional larger steps. The role of Levy statistics in animal behaviour is much debated — as explained in an accompanying News Feature — but the idea of extending it to human behaviour was boosted by a report in 2006 of Levy flight-like patterns in human movement tracked via dollar bills. A new human study, based on tracking the trajectory of 100,000 cell-phone users for six months, reveals behaviour close to a Levy pattern, but deviating from it as individual trajectories show a high degree of temporal and spatial regularity: work and other commitments mean we are not as free to roam as a foraging animal. But by correcting the data to accommodate individual variation, simple and predictable patterns in human travel begin to emerge. The cover photo (by Cesar Hidalgo) captures human mobility in New York's Grand Central Station. This study used a sample of 100,000 mobile phone users whose trajectory was tracked for six months to study human mobility patterns. Displacements across all users suggest behaviour close to the Levy-flight-like pattern observed previously based on the motion of marked dollar bills, but with a cutoff in the distribution. The origin of the Levy patterns observed in the aggregate data appears to be population heterogeneity and not Levy patterns at the level of the individual. Despite their importance for urban planning1, traffic forecasting2 and the spread of biological3,4,5 and mobile viruses6, our understanding of the basic laws governing human motion remains limited owing to the lack of tools to monitor the time-resolved location of individuals. Here we study the trajectory of 100,000 anonymized mobile phone users whose position is tracked for a six-month period. We find that, in contrast with the random trajectories predicted by the prevailing Levy flight and random walk models7, human trajectories show a high degree of temporal and spatial regularity, each individual being characterized by a time-independent characteristic travel distance and a significant probability to return to a few highly frequented locations. After correcting for differences in travel distances and the inherent anisotropy of each trajectory, the individual travel patterns collapse into a single spatial probability distribution, indicating that, despite the diversity of their travel history, humans follow simple reproducible patterns. This inherent similarity in travel patterns could impact all phenomena driven by human mobility, from epidemic prevention to emergency response, urban planning and agent-based modelling.

/pdf/understanding-individual-human-mobility-patterns-3k1tyob274.pdf

Understanding individual human mobility patterns

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

We conduct a comparative experimental analysis of three well known media access protocols: 802.11, CSMA, and MACA for wireless radio networks. Both fixed and ad-hoc networks are considered. The experimental analysis was carried out using GloMoSim: a tool for simulating wireless networks. The main focus of experiments was to study how (i) the size of the network, (ii) number of open connections, (iii) the spatial location of individual connections, (iv) speed with which individual nodes move and (v) protocols higher up in the protocol stack (e,g. routing layer) affect the performance of the media access sublayer protocols. The performance of the protocols was measured w.r.t. three important parameters: (1) number of received packets, (2) average latency of each packet, and (3) throughput. The following general qualitative conclusions were obtained; some of the conclusions reinforce the earlier claims by other researchers. (1) Although 802.11 performs better than the other two protocols with respect to fairness of transmission, packets dropped, and latency, its performance is found to (i) show a lot of variance with changing input parameters and (ii) the overall performance still leaves a lot of room for improvement. (2) CSMA does not perform too well under the fairness criteria, however, was the best in terms of the latency criteria. (3) MACA also shows fairness problems and has poor performance at high packet injection rates. (4) Protocols in the higher level of the protocol stack affect the MAC layer performance. The main general implications of our work is two folds: (1) No single protocol dominated the other protocols across various measures of efficiency. This motivates the design of a new class of parameterized protocols that adapt to changes in the network connectivity and loads. We refer to these class of protocols as parameterized dynamically adaptive efficient protocols and as a first step suggest key design requirements for such a class of protocols. (2) Performance analysis of protocols at a given level in the protocol stack need to be studied not locally in isolation but as a part of the complete protocol stack. The results suggest that in order to improve the performance of a communication network, it will be important to study the entire protocol stuck as a single algorithmic construct; optimizing individual layers in the 7 layer OSI stack will not yield performance improvements beyond a point.

/pdf/a-comparative-experimental-study-of-media-access-protocols-1jukrherxa.pdf

A Comparative experimental study of media access protocols for wireless radio networks

The authors present a simple extensible theoretical framework for devising polynomial time approximation schemes for problems represented using natural syntactic (algebraic) specifications endowed with natural graph theoretic restrictions on input instances. Direct application of the technique yields polynomial time approximation schemes for all the problems studied in [LT80, NC88, KM96, Ba83, DTS93, HM+94a, HM+94] as well as the first known approximation schemes for a number of additional combinatorial problems. One notable aspect of the work is that it provides insights into the structure of the syntactic specifications and the corresponding algorithms considered in [KM96, HM+94]. The understanding allows them to extend the class of syntactic specifications for which generic approximation schemes can be developed. The results can be shown to be tight in many cases, i.e. natural extensions of the specifications can be shown to yield non-approximable problems. The results provide a non-trivial characterization of a class of problems having a PTAS and extend the earlier work on this topic by [KM96, HM+94].

/pdf/towards-syntactic-characterizations-of-approximation-schemes-6m5vnrdztg.pdf

Towards syntactic characterizations of approximation schemes via predicate and graph decompositions

Several practical instances of network design problems often require the network to satisfy multiple constraints. In this paper, we focus on the following problem (and its variants): find a low-cost network, under one cost function, that services every node in the graph, under another cost function, (i.e., every node of the graph is Within a prespecified distance from the network). Such problems find applications in optical network design and the efficient maintenance of distributed databases.We utilize the framework developed in Marathe et al. [1995] co formulate these problems as bicriteria network design problems, and present approximation algorithms for a class of service-constrained network design problems. We also present lower bounds on the approximability of these problems that demonstrate that our performance ratios are close to best possible.

Service-constrained network design problems

Motivated by applications in cartography and computer graphics, we study a version of the map-labeling problem that we call the k-Position Map-Labeling Problem: given a set of points in the plane and, for each point, a set of up to k allowable positions, place uniform and non-intersecting labels of maximum size at each point in one of the allowable positions. This version combines an aesthetic criterion and a legibility criterion and comes close to actual practice while generalizing the fixed-point and slider models found in the literature. We present a general heuristic that given an ∊ > 0, runs in time O(n log n + n log(R*/ ∊) log(k)), where R* is the size of the optimal label, and guarantees a constant approximation for any regular labels. For circular labels, our technique yields a (3.6 + ∊)-approximation, improving in the case of arbitrary placement over the previous bound of approximately 19.5 obtained by Strijk and Wolff.28 We then extend our approach to arbitrary positions, obtaining an algorithm that is easy to implement and also substantially improves the best approximation bounds. Our technique combines several geometric and combinatorial properties, which may be of independent interest.

/pdf/point-set-labeling-with-specified-positions-1b8mpzt479.pdf

Point set labeling with specified positions

We characterize the complexity of several basic optimization problems for unit disk graphs specified hierarchically as in [LW87a, Le88, LW92]. Both PSPACE-hardness results and polynomial time approximations are presented for most of the problems considered. These problems include minimum vertex coloring, maximum independent set, minimum clique cover, minimum dominating set and minimum independent dominating set.

Madhav V. Marathe

Papers

A Comparative experimental study of media access protocols for wireless radio networks

Towards syntactic characterizations of approximation schemes via predicate and graph decompositions

Service-constrained network design problems

Point set labeling with specified positions

Hierarchical Specified Unit Disk Graphs (Extended Abstract)