Lasso is a regularization method for parameter estimation in linear models. It optimizes the model parameters with respect to a loss function subject to model complexities. This paper explores the use of lasso for statistical language modeling for text input. Owing to the very large number of parameters, directly optimizing the penalized lasso loss function is impossible.
The modelling of mechanical systems provides engineers and students with the methods to model and understand mechanical systems by using both mathematical and computer-based tools. Written by an eminent authority in the field, this is the second of four volumes which provide engineers with a comprehensive resource on this cornerstone mechanical engineering subject.
Dealing with continuous systems, this book covers solid mechanics, beams, plates and shells.
Monte Carlo methods are ubiquitous in applications in the finance and
insurance industry. They are often the only accessible tool for financial engineers
and actuaries when it comes to complicated price or risk computations,
in particular for those that are based on many underlyings. However, as they
tend to be slow, it is very important to have a big tool box for speeding them
up or – equivalently – for increasing their accuracy. Further, recent years have
seen a lot of developments in Monte Carlo methods with a high potential for
success in applications.
Chris Adcock is Professor of Financial Econometrics in the University of Sheffield. His
career includes several years working in quantitative investment management in the
City and, prior to that, a decade in management science consultancy. His research
interests are in the development of robust and non-standard methods for modelling
expected returns, portfolio selection methods and the properties of optimized portfolios.
He has acted as an advisor to a number of asset management firms. He is the
founding editor of the European Journal of Finance.
This Second Edition of the go-to reference combines the classical analysis and modern applications of applied mathematics for chemical engineers. The book introduces traditional techniques for solving ordinary differential equations (ODEs), adding new material on approximate solution methods such as perturbation techniques and elementary numerical solutions. It also includes analytical methods to deal with important classes of finite-difference equations. The last half discusses numerical solution techniques and partial differential equations (PDEs). The read...
This book was conceived as a result of many years research with students
and postdocs in molecular simulation, and shaped over several courses on
the subject given at the University of Groningen, the Eidgen¨ossische Technische
Hochschule (ETH) in Z¨urich, the University of Cambridge, UK, the
University of Rome (La Sapienza), and the University of North Carolina
at Chapel Hill, NC, USA.
The goal of this book is to expose the reader to modern computational tools for
solving differential equation models that arise in chemical engineering, e.g.,
diffusion-reaction, mass-heat transfer, and fluid flow. The emphasis is placed
on the understanding and proper use of software packages. In each chapter we
outline numerical techniques that either illustrate a computational property of
interest or are the underlying methods of a computer package. At the close of
each chapter a survey of computer packages is accompanied by examples of
This paper presents a method to develop a class of variable memory Markov models that have higher memory capacity than traditional (uniform memory) Markov models. The structure of the variable memory models is induced from a manually annotated corpus through a decision tree learning algorithm. A series of comparative experiments show the resulting models outperform uniform memory Markov models in a part-of-speech tagging task.
Statistical approaches to automatic text summarization based on term frequency continue to perform on par with more complex summarization methods. To compute useful frequency statistics, however, the semantically important words must be separated from the low-content function words. The standard approach of using an a priori stopword list tends to result in both undercoverage, where syntactical words are seen as semantically relevant, and overcoverage, where words related to content are ignored. ...
We study the issue of porting a known NLP method to a language with little existing NLP resources, speciﬁcally Hebrew SVM-based chunking. We introduce two SVM-based methods – Model Tampering and Anchored Learning. These allow ﬁne grained analysis of the learned SVM models, which provides guidance to identify errors in the training corpus, distinguish the role and interaction of lexical features and eventually construct a model with ∼10% error reduction.
We investigate a number of simple methods for improving the word-alignment accuracy of IBM Model 1. We demonstrate reduction in alignment error rate of approximately 30% resulting from (1) giving extra weight to the probability of alignment to the null word, (2) smoothing probability estimates for rare words, and (3) using a simple heuristic estimation method to initialize, or replace, EM training of model parameters.
In Chinese, a word is made up of one or more characters. Hence, there also exists preferred relationships between Chinese characters. [Sproat+90] employed a statistical method to group neighboring Chinese characters in a sentence into two-character words by making use of a measure of character association based on mutual information. Here, we will focus instead on the preferred relationships between words. The preference relationships between words can expand from a short to long distance.
This paper presents a method to combine a set of unsupervised algorithms that can accurately disambiguate word senses in a large, completely untagged corpus. Although most of the techniques for word sense resolution have been presented as stand-alone, it is our belief that full-fledged lexical ambiguity resolution should combine several information sources and techniques. The set of techniques have been applied in a combined way to disambiguate the genus terms of two machine-readable dictionaries (MRD), enabling us to construct complete taxonomies for Spanish and French. ...
The computation of selectional preferences, the admissible argument values for a relation, is a well-known NLP task with broad applicability. We present L DA - SP, which utilizes LinkLDA (Erosheva et al., 2004) to model selectional preferences. By simultaneously inferring latent topics and topic distributions over relations, L DA - SP combines the beneﬁts of previous approaches: like traditional classbased approaches, it produces humaninterpretable classes describing each relation’s preferences, but it is competitive with non-class-based methods in predictive power. ...
This paper describes the application of so-called topic models to selectional preference induction. Three models related to Latent Dirichlet Allocation, a proven method for modelling document-word cooccurrences, are presented and evaluated on datasets of human plausibility judgements. Compared to previously proposed techniques, these models perform very competitively, especially for infrequent predicate-argument combinations where they exceed the quality of Web-scale predictions while using relatively little data. ...
This paper presents a joint optimization method of a two-step conditional random ﬁeld (CRF) model for machine transliteration and a fast decoding algorithm for the proposed method. Our method lies in the category of direct orthographical mapping (DOM) between two languages without using any intermediate phonemic mapping. In the two-step CRF model, the ﬁrst CRF segments an input word into chunks and the second one converts each chunk into one unit in the target language. In this paper, we propose a method to jointly optimize the two-step CRFs and also a fast algorithm to realize it. ...
In this paper, we present a reinforcement learning approach for mapping natural language instructions to sequences of executable actions. We assume access to a reward function that deﬁnes the quality of the executed actions. During training, the learner repeatedly constructs action sequences for a set of documents, executes those actions, and observes the resulting reward. We use a policy gradient algorithm to estimate the parameters of a log-linear model for action selection.
The modules ‘IT as an Enabler…’, ‘Methods and Models of Business Engineering’, ‘Information Man-
agement’ and ‘Business Models of the Information Age’ constitute the information systems view on
Being potential enablers for new business models and reinvented processes, IT innovations and novel
standardized software packages as well as appropriate information systems architectures are addressed.
Also more technical topics like media convergence, platforms for electronic commerce, infrastru ctures
for mobile work etc. are cove red.