7,806 research outputs found
Prediction of relevance of an image from a scan pattern
No full textThis report considers the task of inferring implicit relevance feedback from eye movements in image retrieval settings. The feasibility of solving the problem without using any image-level features is demonstrated on two different search settings, and the accuracy of inferring the relevance feedback is shown to be relatively high, clearly better than random. In addition, the report provides a list of image-level features that are good cues for relevance
Samuel Dorris Dickinson papers
No full textThe Samuel Dorris Dickinson papers contain the professional and personal records of archaeologist, journalist, and author Samuel Dorris Dickinson
Predicting relevance of parts of an image
No full textThis report studies the task of inferring which parts of an image are relevant for the user viewing the image. The relevance is inferred from gaze trajectory of users viewing the images given a specific task. Novel computational models based on both Bayesian generative modeling and kernel methods are developed for inferring the regions of interest from raw fixation data, as well as from combination of eye movements and image content features
Ranking algorithms for implicit feedback
No full textThis report presents novel algorithms to use eye movements as an implicit relevance feedback in order to improve the performance of the searches. The algorithms are evaluated on "Transport Rank Five" Dataset which were previously collected in Task 8.3. We demonstrated that simple linear combination or tensor product of eye movement and image features can improve the retrieval accuracy
Learning to Rank Images from Eye Movements
Full text linkCombining multiple information sources can improve the accuracy of search in information retrieval. This paper presents a new image search strategy which combines image features together with implicit feedback from users' eye movements, using them to rank images. In order to better deal with larger data sets, we present a perceptron formulation of the Ranking Support Vector Machine algorithm. We present initial results on inferring the rank of images presented in a page based on simple image features and implicit feedback of users. The results show that the perceptron algorithm improves the results, and that fusing eye movements and image histograms gives better rankings to images than either of these features alone
Portrait of author David Foster at the National Library of Australia, Canberra, 8 June 2011 /
No full textTitle from acquisitions documentation.; Part of the collection: Portraits of author David Foster at the National Library of Australia, Canberra, 8 June 2011.; Acquired in digital format; access copy available online.; Mode of access: Online.; Photographed by a staff member of the National Library of Australia
Author David Foster with academic Jeff Doyle at the National Library of Australia, Canberra, 8 June 2011 /
No full textTitle from acquisitions documentation.; Part of the collection: Portraits of author David Foster at the National Library of Australia, Canberra, 8 June 2011.; Acquired in digital format; access copy available online.; Mode of access: Online.; Photographed by a staff member of the National Library of Australia
Author David Foster and academic Jeff Doyle at the National Library of Australia, Canberra, 8 June 2011 /
No full textTitle from acquisitions documentation.; Part of the collection: Portraits of author David Foster at the National Library of Australia, Canberra, 8 June 2011.; Acquired in digital format; access copy available online.; Mode of access: Online.; Photographed by a staff member of the National Library of Australia
Portrait of Paul Ham at the National Library of Australia, 15 November 2011 /
No full textTitle from nformation supplied by photographer.; Part of the collection: Podcast photograph of author Paul Ham at the National Library of Australia, 15 November 2011.; Mode of access: Online.; Photographed by a staff member of the National Library of Australia
Molecular property prediction using pretrained-BERT and Bayesian active learning : a data-efficient approach to drug design
No full textPublisher Copyright: © The Author(s) 2025. | openaire: EC/H2020/956832/EU//AIDDAbstract: In drug discovery, prioritizing compounds for experimental testing is a critical task that can be optimized through active learning by strategically selecting informative molecules. Active learning typically trains models on labeled examples alone, while unlabeled data is only used for acquisition. This fully supervised approach neglects valuable information present in unlabeled molecular data, impairing both predictive performance and the molecule selection process. We address this limitation by integrating a transformer-based BERT model, pretrained on 1.26 million compounds, into the active learning pipeline. This effectively disentangles representation learning and uncertainty estimation, leading to more reliable molecule selection. Experiments on Tox21 and ClinTox datasets demonstrate that our approach achieves equivalent toxic compound identification with 50% fewer iterations compared to conventional active learning. Analysis reveals that pretrained BERT representations generate a structured embedding space enabling reliable uncertainty estimation despite limited labeled data, confirmed through Expected Calibration Error measurements. This work establishes that combining pretrained molecular representations with active learning significantly improves both model performance and acquisition efficiency in drug discovery, providing a scalable framework for compound prioritization. Scientific Contribution: We demonstrate that high-quality molecular representations fundamentally determine active learning success in drug discovery, outweighing acquisition strategy selection. We provide a framework that integrates pretrained transformer models with Bayesian active learning to separate representation learning from uncertainty estimation—a critical distinction in low-data scenarios. This approach establishes a foundation for more efficient screening workflows across diverse pharmaceutical applications.Peer reviewe
- …
