Competitive analysis of incentive compatible on-line auctions

This paper studies auctions in a setting where the different bidders arrive at different times and the auction mechanism is required to make decisions about each bid as it is received. Such settings occur in computerized auctions of computational resources as well as in other settings. We call such auctions, on-line auctions.

We first characterize exactly on-line auctions that are incentive compatible, i.e. where rational bidders are always motivated to bid their true valuation. We then embark on a competitive worst-case analysis of incentive compatible on-line auctions. We obtain several results, the cleanest of which is an incentive compatible on-line auction for a large number of identical items. This auction has an optimal competitive ratio, both in terms of seller's revenue and in terms of the total social efficiency obtained.     

Lessons and Challenges from Mining Retail E-Commerce Data

The architecture of Blue Martini Software's e-commerce suite has supported data collection, data transformation, and data mining since its inception. With clickstreams being collected at the application-server layer, high-level events being logged, and data automatically transformed into a data warehouse using meta-data, common problems plaguing data mining using weblogs (e.g., sessionization and conflating multi-sourced data) were obviated, thus allowing us to concentrate on actual data mining goals. The paper briefly reviews the architecture and discusses many lessons learned over the last four years and the challenges that still need to be addressed. The lessons and challenges are presented across two dimensions: business-level vs. technical, and throughout the data mining lifecycle stages of data collection, data warehouse construction, business intelligence, and deployment. The lessons and challenges are also widely applicable to data mining domains outside retail e-commerce.     

TRADE‐OFF BETWEEN APPROXIMATION ACCURACY AND COMPLEXITY FOR TS FUZZY MODELS

This paper proposes a transformation method that serves the trade‐off between the modelling complexity and accuracy of multi‐variable Takagi‐Sugeno fuzzy inference operator‐based modelling (TS fuzzy modelling). The relation between the number of fuzzy rules and the modelling accuracy is defined in the paper. The proposed transformation method is capable of finding the minimal number of fuzzy rules for a given accuracy of a given TS fuzzy model. A case study, focusing on a benchmark problem of fault diagnosis, developed in the framework of EC‐founded Research Training Network DAMADICS, of an actuator in a sugar factory, is presented to provide feasibility of the proposed method.     

Thyroid cancer after exposure to external radiation: a pooled analysis of seven studies

The thyroid gland of children is especially vulnerable to the carcinogenic action of ionizing radiation. To provide insights into various modifying influences on risk, seven major studies with organ doses to individual subjects were evaluated. Five cohort studies (atomic bomb survivors, children treated for tinea capitis, two studies of children irradiated for enlarged tonsils, and infants irradiated for an enlarged thymus gland) and two case-control studies (patients with cervical cancer and childhood cancer) were studied. The combined studies include almost 120,000 people (approximately 58,000 exposed to a wide range of doses and 61,000 nonexposed subjects), nearly 700 thyroid cancers and 3,000,000 person years of follow-up. For persons exposed to radiation before age 15 years, linearity best described the dose response, even down to 0.10 Gy. At the highest doses (> 10 Gy), associated with cancer therapy, there appeared to be a decrease or leveling of risk. For childhood exposures, the pooled excess relative risk per Gy (ERR/Gy) was 7.7 (95% CI = 2.1, 28.7) and the excess absolute risk per 10(4) PY Gy (EAR/10(4) PY Gy) was 4.4 (95% CI = 1.9, 10.1). The attributable risk percent (AR%) at 1 Gy was 88%. However, these summary estimates were affected strongly by age at exposure even within this limited age range. The ERR was greater (P = 0.07) for females than males, but the findings from the individual studies were not consistent. The EAR was higher among women, reflecting their higher rate of naturally occurring thyroid cancer. The distribution of ERR over time followed neither a simple multiplicative nor an additive pattern in relation to background occurrence. Only two cases were seen within 5 years of exposure. The ERR began to decline about 30 years after exposure but was still elevated at 40 years. Risk also decreased significantly with increasing age at exposure, with little risk apparent after age 20 years. Based on limited data, there was a suggestion that spreading dose over time (from a few days to > 1 year) may lower risk, possibly due to the opportunity for cellular repair mechanisms to operate. The thyroid gland in children has one of the highest risk coefficients of any organ and is the only tissue with convincing evidence for risk about 1.10 Gy.     

Speech separation using speaker-adapted eigenvoice speech models

We present a system for model-based source separation for use on single channel speech mixtures where the precise source characteristics are not known a priori. The sources are modeled using hidden Markov models (HMM) and separated using factorial HMM methods. Without prior speaker models for the sources in the mixture it is difficult to exactly resolve the individual sources because there is no way to determine which state corresponds to which source at any point in time. This is solved to a small extent by the temporal constraints provided by the Markov models, but permutations between sources remains a significant problem. We overcome this by adapting the models to match the sources in the mixture. We do this by representing the space of speaker variation with a parametric signal model-based on the eigenvoice technique for rapid speaker adaptation. We present an algorithm to infer the characteristics of the sources present in a mixture, allowing for significantly improved separation performance over that obtained using unadapted source models. The algorithm is evaluated on the task defined in the 2006 Speech Separation Challenge [Cooke, M.P., Lee, T.-W., 2008. The 2006 Speech Separation Challenge. Computer Speech and Language] and compared with separation using source-dependent models. Although performance is not as good as with speaker-dependent models, we show that the system based on model adaptation is able to generalize better to held out speakers.     

Topology-Invariant Similarity of Nonrigid Shapes

This paper explores the problem of similarity criteria between nonrigid shapes. Broadly speaking, such criteria are divided into intrinsic and extrinsic, the first referring to the metric structure of the object and the latter to how it is laid out in the Euclidean space. Both criteria have their advantages and disadvantages: extrinsic similarity is sensitive to nonrigid deformations, while intrinsic similarity is sensitive to topological noise. In this paper, we approach the problem from the perspective of metric geometry. We show that by unifying the extrinsic and intrinsic similarity criteria, it is possible to obtain a stronger topology-invariant similarity, suitable for comparing deformed shapes with different topology. We construct this new joint criterion as a tradeoff between the extrinsic and intrinsic similarity and use it as a set-valued distance. Numerical results demonstrate the efficiency of our approach in cases where using either extrinsic or intrinsic criteria alone would fail.     

Sublinear Algorithms for Approximating String Compressibility

We raise the question of approximating the compressibility of a string with respect to a fixed compression scheme, in sublinear time. We study this question in detail for two popular lossless compression schemes: run-length encoding (RLE) and a variant of Lempel-Ziv (LZ77), and present sublinear algorithms for approximating compressibility with respect to both schemes. We also give several lower bounds that show that our algorithms for both schemes cannot be improved significantly. Our investigation of LZ77 yields results whose interest goes beyond the initial questions we set out to study. In particular, we prove combinatorial structural lemmas that relate the compressibility of a string with respect to LZ77 to the number of distinct short substrings contained in it (its ?th subword complexity , for small ?). In addition, we show that approximating the compressibility with respect to LZ77 is related to approximating the support size of a distribution.     

数字合成(上)

什么是数字合成?当我们讨论它的时候,首先我们试着把多个元素整合成一个单个的完整的整体图像,当大多数这些技术仍然适用于图像时,我们将关注有用的和合理的能处理大的图像序列的工具和方法。正如您所见的,一个技术很好的合成师是可以包括从技师到艺术家的,不只是需要基本了解“操作的工具”,这里包括多种软件,而且必须是一个能洞察视觉本质的人。记住这一点,当遇到“它看起来是否正确?”的问题时,所有技术考虑都是不重要了,很明显只是一个对主题的把握,同时一个好的合成师都应以高要求来做出每个决定。定义什么是合成?有一个基本的定义:把…     

Why is a computational framework for motivational and metacognitive control needed?

This paper discusses, in the context of computational modelling and simulation of cognition, the relevance of deeper structures in the control of behaviour. Such deeper structures include motivational control of behaviour, which provides underlying causes for actions, and also metacognitive control, which provides higher-order processes for monitoring and regulation. It is argued that such deeper structures are important and thus cannot be ignored in computational cognitive architectures. A general framework based on the Clarion cognitive architecture is outlined that emphasises the interaction amongst action selection, motivation, and metacognition. The upshot is that it is necessary to incorporate all essential processes; short of that, the understanding of cognition can only be incomplete.