Multiple Instance Learning For Image Search

Multiple Instance Learning for Image Search

Image and Video Retrieval

It was our great pleasure to host the 4th International Conference on Image and Video Retrieval (CIVR) at the National University of Singapore on 20–22 July 2005. CIVR aims to provide an international forum for the discussion of research challenges and exchange of ideas among researchers and practitioners in image/video retrieval technologies. It addresses innovative research in the broad ?eld of image and video retrieval. A unique feature of this conference is the high level of participation by researchers from both academia and industry. Another unique feature of CIVR this year was in its format – it o?ered both the traditional oral presentation sessions, as well as the short presentation cum poster sessions. The latter provided an informal alternative forum for animated discussions and exchanges of ideas among the participants. We are pleased to note that interest in CIVR has grown over the years. The number of submissions has steadily increased from 82 in 2002, to 119 in 2003, and 125 in 2004. This year, we received 128 submissions from the international communities:with81(63.3%)fromAsiaandAustralia,25(19.5%)fromEurope, and 22 (17.2%) from North America. After a rigorous review process, 20 papers were accepted for oral presentations, and 42 papers were accepted for poster presentations. In addition to the accepted submitted papers, the program also included 4 invited papers, 1 keynote industrial paper, and 4 invited industrial papers. Altogether, we o?ered a diverse and interesting program, addressing the current interests and future trends in this area.

Introduction to Semi-Supervised Learning

Semi-supervised learning is a learning paradigm concerned with the study of how computers and natural systems such as humans learn in the presence of both labeled and unlabeled data. Traditionally, learning has been studied either in the unsupervised paradigm (e.g., clustering, outlier detection) where all the data are unlabeled, or in the supervised paradigm (e.g., classification, regression) where all the data are labeled. The goal of semi-supervised learning is to understand how combining labeled and unlabeled data may change the learning behavior, and design algorithms that take advantage of such a combination. Semi-supervised learning is of great interest in machine learning and data mining because it can use readily available unlabeled data to improve supervised learning tasks when the labeled data are scarce or expensive. Semi-supervised learning also shows potential as a quantitative tool to understand human category learning, where most of the input is self-evidently unlabeled. In this introductory book, we present some popular semi-supervised learning models, including self-training, mixture models, co-training and multiview learning, graph-based methods, and semi-supervised support vector machines. For each model, we discuss its basic mathematical formulation. The success of semi-supervised learning depends critically on some underlying assumptions. We emphasize the assumptions made by each model and give counterexamples when appropriate to demonstrate the limitations of the different models. In addition, we discuss semi-supervised learning for cognitive psychology. Finally, we give a computational learning theoretic perspective on semi-supervised learning, and we conclude the book with a brief discussion of open questions in the field. Table of Contents: Introduction to Statistical Machine Learning / Overview of Semi-Supervised Learning / Mixture Models and EM / Co-Training / Graph-Based Semi-Supervised Learning / Semi-Supervised Support Vector Machines / Human Semi-Supervised Learning / Theory and Outlook