Harvard University

About: Harvard University is a education organization based out in Cambridge, Massachusetts, United States. It is known for research contribution in the topics: Population & Cancer. The organization has 208150 authors who have published 530388 publications receiving 38152182 citations. The organization is also known as: Harvard & University of Harvard.

Traumatic arthritis of the hip after dislocation and acetabular fractures: treatment by mold arthroplasty. An end-result study using a new method of result evaluation

Abstract: An end-result analysis is presented of thirty-nine mold arthroplasties performed at the Massachusetts General Hospital between 1945 and 1965 in thirty-eight consecutive private patients for arthritis of the hip following fractures of the acetabulum or dislocations of the hip. Of the nineteen unilateral cases in the second half of the series, sixteen were rated good or excellent. Results in the second half of the series were significantly better statistically than those in the first half of the series. Possible reasons for this improvement are discussed. No significant deterioration occurred with the passage of time. Among the thirty-nine hips, three revisions were required. One patient had postoperative sepsis after arthroplasty. Four patients who had had intra-articular sepsis prior to arthroplasty showed no evidence of sepsis postoperatively. Factors influencing the choice between hip fusion and hip arthroplasty in these cases are presented. A new system for rating hip function is proposed and is compared with the systems of Larson and Shepherd.

Practical Bayesian Optimization of Machine Learning Algorithms

Abstract: The use of machine learning algorithms frequently involves careful tuning of learning parameters and model hyperparameters. Unfortunately, this tuning is often a "black art" requiring expert experience, rules of thumb, or sometimes brute-force search. There is therefore great appeal for automatic approaches that can optimize the performance of any given learning algorithm to the problem at hand. In this work, we consider this problem through the framework of Bayesian optimization, in which a learning algorithm's generalization performance is modeled as a sample from a Gaussian process (GP). We show that certain choices for the nature of the GP, such as the type of kernel and the treatment of its hyperparameters, can play a crucial role in obtaining a good optimizer that can achieve expertlevel performance. We describe new algorithms that take into account the variable cost (duration) of learning algorithm experiments and that can leverage the presence of multiple cores for parallel experimentation. We show that these proposed algorithms improve on previous automatic procedures and can reach or surpass human expert-level optimization for many algorithms including latent Dirichlet allocation, structured SVMs and convolutional neural networks.

Increasing Returns, Path Dependence, and the Study of Politics

Abstract: It is increasingly common for social scientists to describe political processes as “path dependent.” The concept, however, is often employed without careful elaboration. This article conceptualizes path dependence as a social process grounded in a dynamic of “increasing returns.” Reviewing recent literature in economics and suggesting extensions to the world of politics, the article demonstrates that increasing returns processes are likely to be prevalent, and that good analytical foundations exist for exploring their causes and consequences. The investigation of increasing returns can provide a more rigorous framework for developing some of the key claims of recent scholarship in historical institutionalism: Specific patterns of timing and sequence matter; a wide range of social outcomes may be possible; large consequences may result from relatively small or contingent events; particular courses of action, once introduced, can be almost impossible to reverse; and consequently, political development is punctuated by critical moments or junctures that shape the basic contours of social life.

The Structure of Haplotype Blocks in the Human Genome

Abstract: Haplotype-based methods offer a powerful approach to disease gene mapping, based on the association between causal mutations and the ancestral haplotypes on which they arose. As part of The SNP Consortium Allele Frequency Projects, we characterized haplotype patterns across 51 autosomal regions (spanning 13 megabases of the human genome) in samples from Africa, Europe, and Asia. We show that the human genome can be parsed objectively into haplotype blocks: sizable regions over which there is little evidence for historical recombination and within which only a few common haplotypes are observed. The boundaries of blocks and specific haplotypes they contain are highly correlated across populations. We demonstrate that such haplotype frameworks provide substantial statistical power in association studies of common genetic variation across each region. Our results provide a foundation for the construction of a haplotype map of the human genome, facilitating comprehensive genetic association studies of human disease.