Results 101 to 110 of about 842,651 (374)
Prompt-based learning for few-shot class-incremental learning
Alexandria Engineering JournalFew-Shot Class-Incremental Learning (FSCIL) aims to enable deep neural networks to incrementally learn new tasks from a limited number of labeled samples, while retaining knowledge of previously learned tasks, mimicking the way humans learn.
Jicheng Yuan +6 more
doaj +1 more source
IEEE Access, 2020 State-of-the-art deep learning models for food recognition do not allow data incremental learning and often suffer from catastrophic interference problems during the class incremental learning.
Ghalib Ahmed Tahir, Chu Kiong Loo
doaj +1 more source
Incremental class representation learning for face recognition [PDF]
, 2017 Image classification is one of the most active challenging problems in computer vision field. Taking this to Deep Neural Networks with systems that are able to deal with large data sets as it can be ImageNet.
Presas Valga, Eric
core
Class-incremental Learning with Pre-allocated Fixed Classifiers [PDF]
, 2021 Federico Pernici +4 more
openalex +1 more source
Class Incremental Learning for Algorithm Selection
Proceedings of the Genetic and Evolutionary Computation Conference CompanionAlgorithm selection is commonly used to predict the best solver from a portfolio per per-instance. In many real scenarios, instances arrive in a stream: new instances become available over time, while the number of class labels can also grow as new data distributions arrive downstream.
Mate Botond Nemeth +3 more
openaire +2 more sources
A Lightweight Procedural Layer for Hybrid Experimental–Computational Workflows in Materials Science
Advanced Engineering Materials, EarlyView.We unveil a prototype hybrid‐workflow framework that fuses automatedcomputation with hands‐on experiments. Built atop pyiron, a lightweight, parameterized layer translates procedure descriptions into executable manual steps, syncing instrument settings, human interventions, and data capture in real‐time today.
Steffen Brinckmann +8 more
wiley +1 more source
Applied SciencesEvent detection is a crucial task in information extraction. Existing research primarily focuses on machine automatic detection tasks, which often perform poorly in certain practical applications.
Jiashun Duan, Xin Zhang
doaj +1 more source
Open Set Recognition Based on Meta Class Incremental Learning [PDF]
Jisuanji kexueTraditional image classification algorithms assume that the world is static and closed,whereas the real world is dyna-mic and open,and new categories and their samples are continually emerging,leading to a decrease in the accuracies of traditional image ...
SUN Jinyong, WANG Xuechun, CAI Guoyong, SHANG Zhiliang
doaj +1 more source
PASTA‐ELN: Simplifying Research Data Management for Experimental Materials Science
Advanced Engineering Materials, EarlyView.Research data management faces ongoing hurdles as many ELNs remain complex and restrictive. PASTA‐ELN offers an open‐source, cross‐platform solution that prioritizes simplicity, offline access, and user control. Its in tuitive folder structure, modular Python add‐ons, and open formats enable seamless documentation, FAIR data practices, and easy ...
S. Brinckmann, G. Winkens, R. Schwaiger
wiley +1 more source
IEEE Access, 2019 With the development of deep convolutional neural networks in recent years, the network structure has become more and more complicated and varied, and there are very good results in pattern recognition, image classification, scene classification, and ...
Xin Ye, Qiuyu Zhu
doaj +1 more source