AI Article Synopsis
- Conventional active learning methods for video/image retrieval struggle with unknown query distributions, leading to challenges in balancing exploration and exploitation of the feature space.
- The paper introduces a new method called coached active learning, which predicts the query distribution through training, allowing for more informed searching.
- Experiments conducted on TRECVID 2005-2009 data show that this approach is efficient, effective, and outperforms traditional active learning methods and several advanced interactive video retrieval systems.
Article Abstract
Conventional active learning approaches for interactive video/image retrieval usually assume the query distribution is unknown, as it is difficult to estimate with only a limited number of labeled instances available. Thus, it is easy to put the system in a dilemma whether to explore the feature space in uncertain areas for a better understanding of the query distribution or to harvest in certain areas for more relevant instances. In this paper, we propose a novel approach called coached active learning that makes the query distribution predictable through training and, therefore, avoids the risk of searching on a completely unknown space. The estimated distribution, which provides a more global view of the feature space, can be used to schedule not only the timing but also the step sizes of the exploration and the exploitation in a principled way. The results of the experiments on a large-scale data set from TRECVID 2005-2009 validate the efficiency and effectiveness of our approach, which demonstrates an encouraging performance when facing domain-shift, outperforms eight conventional active learning methods, and shows superiority to six state-of-the-art interactive video retrieval systems.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/TIP.2012.2222902 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Background: Manual analysis of histopathological images is often not only time-consuming and painstaking but also prone to error from subjective evaluation criteria and human error. To address these issues, we created a fully automated workflow to enumerate jejunal crypts in a microcolony survival assay to quantify gastrointestinal damage from radiation.
Methods And Materials: After abdominal irradiation of mice, jejuna were obtained and prepared on histopathologic slides, and crypts were counted manually by trained individuals.
How are associative memories formed? Which cells represent a memory, and when are they engaged? By visualizing and tagging cells based on their calcium influx with unparalleled temporal precision, we identified non-overlapping dorsal CA1 neuronal ensembles that are differentially active during associative fear memory acquisition. We dissected the acquisition experience into periods during which salient stimuli were presented or certain mouse behaviors occurred and found that cells associated with specific acquisition periods are sufficient alone to drive memory expression and contribute to fear engram formation. This study delineated the different identities of the cell ensembles active during learning, and revealed, for the first time, which ones form the core engram and are essential for memory formation and recall.
View Article and Find Full Text PDFDevelopment and evaluation of a machine learning model for post-surgical acute kidney injury in active infective endocarditis.
Front Cardiovasc Med
December 2024
Chief of Cardiac Surgery, Peking Union Medical College Hospital, Beijing, China.
Introduction: Acute kidney injury (AKI) is notably prevalent after cardiac surgery for patients with active infective endocarditis. This study aims to create a machine learning model to predict AKI in this high-risk group, improving upon existing models by focusing specifically on endocarditis-related surgeries.
Methods: We analyzed medical records from 527 patients who underwent cardiac surgery for active infective endocarditis from January 2012 to December 2023.
Evaluating Active Learning Strategies for Automated Classification of Patient Safety Event Reports in Hospitals.
Proc Hum Factors Ergon Soc Annu Meet
September 2024
Mechanical and Industrial Engineering, University of Toronto, Toronto, ON, Canada.
Patient safety event (PSE) reports, which document incidents that compromise patient safety, are fundamental for improving healthcare quality. Accurate classification of these reports is crucial for analyzing trends, guiding interventions, and supporting organizational learning. However, this process is labor-intensive due to the high volume and complex taxonomy of reports.
View Article and Find Full Text PDFFlipped classroom-based application of Peyton's four-step approach in standardized training of ultrasound residents for thyroid and cervical lymph node zoning.
PeerJ
December 2024
Department of Ultrasound Medicine, the Second Affiliated Hospital of Anhui Medical University, Hefei, China.
Background: To investigate whether combining the flipped classroom approach with Peyton's four-step method can enhance teaching effectiveness in ultrasound (US) zoning of the thyroid and cervical lymph nodes for standardized residency training.
Methods: A total of 66 resident training students were randomly divided into a control group and an observation group. The control group received traditional teaching methods, including "see one, do one" learning, lecture-based learning (LBL), and case-based learning (CBL).
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!