Results 121 to 130 of about 24,131 (257)

Learning with Preserving for Continual Multitask Learning

Proceedings of the AAAI Conference on Artificial Intelligence
Artificial intelligence systems in critical fields like autonomous driving and medical imaging analysis often continually learn new tasks using a shared stream of input data. For instance, after learning to detect traffic signs, a model may later need to learn to classify traffic lights or different types of vehicles using the same camera feed.
Hanchen David Wang, Siwoo Bae, Zirong Chen, Meiyi Ma   +3 more
openaire   +2 more sources

Ritualized reflection in human dissection: Liminality, ethical negotiation, and professional identity formation among South Korean medical students

Anatomical Sciences Education, EarlyView.
Abstract Human dissection is a foundational component of medical education, yet it places students in profound ethical tension between scientific objectification and respect for human dignity. While prior studies have documented students' emotional responses, the structural transformation of their moral narratives over time, particularly within non ...
Jun‐Ki Lee, Hyoungtae Kim, Hyo Hyun Yoo   +2 more
wiley   +1 more source

Learning to Continually Learn with the Bayesian Principle

CoRR
In the present era of deep learning, continual learning research is mainly focused on mitigating forgetting when training a neural network with stochastic gradient descent on a non-stationary stream of data. On the other hand, in the more classical literature of statistical machine learning, many models have sequential Bayesian update rules that yield ...
Soochan Lee, Hyeonseong Jeon, Jaehyeon Son, Gunhee Kim   +3 more
openaire   +3 more sources

Data, not documents: Moving beyond theories of information‐seeking behavior to advance data discovery

Journal of the Association for Information Science and Technology, Volume 76, Issue 4, Page 649-664, April 2025.
Abstract Many theories of human information behavior (HIB) assume that information objects are in text document format. This paper argues four important HIB theories are insufficient for describing users' search strategies for data because of assumptions about the attributes of objects that users seek.
Anthony J. Million, Jeremy York, Sara Lafia, Libby Hemphill   +3 more
wiley   +1 more source

Continual learning for efficient machine learning

, 2020
Deep learning has enjoyed tremendous success over the last decade, but the training of practically useful deep models remains highly inefficient both in terms of the number of weight updates and training samples. To address one aspect of these issues, this thesis studies the continual learning setting whereby a model utilizes a sequence of tasks ...
openaire   +3 more sources

Affective dimensions in the information behavior of forcibly displaced people: A literature review. An Annual Review of Information Science and Technology (ARIST) paper

Journal of the Association for Information Science and Technology, EarlyView.
Abstract This review analyzed 241 scholarly articles published between 2010 and 2025 in information science venues to examine how affect shapes refugees' information behavior during forced migration and to identify additional contextual factors. It identifies seven affective dimensions: anxiety, shame and stigma, grief and loss, frustration, (mis)trust,
Maja Krtalić, Lilach Alon
wiley   +1 more source

Functional Brain Asymmetry Reveals Heterogeneous Subtypes in Autism Spectrum Disorder

Autism Research, EarlyView.
ABSTRACT Heterogeneity is a critical factor in understanding inter‐individual brain and behavioral variability in autism spectrum disorder (ASD). Since individuals with ASD exhibit atypical communication and social interaction skills closely linked to brain lateralization, this study investigated ASD heterogeneity using an asymmetry index that captures
Chae Yeon Kim, Sunghun Kim, Bo‐yong Park   +2 more
wiley   +1 more source

Operationalizing BioSSbD: A safe‐and‐sustainable‐by‐design framework for biorefineries

Biofuels, Bioproducts and Biorefining, EarlyView.
Abstract Biorefineries are central to the transition toward a circular bioeconomy; however, their increasing scale and technological heterogeneity, and the integration of biological, chemical, and thermochemical processes introduce complex challenges related to safety, sustainability, and operational reliability. Existing Safe‐and‐Sustainable‐by‐Design
Fernando Ramonet
wiley   +1 more source

Diversified Adaptive Stock Selection Using Continual Graph Learning and Ensemble Approach

IEEE Access
Stock selection is essential for portfolio diversification to reduce risks and maximize profits. However, stock selection is challenging owing to the non-stationary nature of stock markets.
Jae-Seung Kim, Sang-Ho Kim, Ki-Hoon Lee
doaj   +1 more source

Continual Learning, Fast and Slow

IEEE Transactions on Pattern Analysis and Machine Intelligence
arXiv admin note: substantial text overlap with arXiv:2110 ...
Quang Pham, Chenghao Liu, Steven C. H. Hoi   +2 more
openaire   +4 more sources