Results 261 to 270 of about 1,148,156 (325)
Some of the next articles are maybe not open access.
Pre-training with Random Orthogonal Projection Image Modeling
International Conference on Learning Representations, 2023Masked Image Modeling (MIM) is a powerful self-supervised strategy for visual pre-training without the use of labels. MIM applies random crops to input images, processes them with an encoder, and then recovers the masked inputs with a decoder, which ...
Maryam Haghighat +3 more
semanticscholar +1 more source
Pretreatments by Orthogonal Projections
2021Orthogonal projections belong to the few calculation tools onto which rely the most popular chemometric methods. They can be found in several calibration and classification methods. However, the term orthogonal projection is more commonly associated to the field of pretreatments (or preprocessings).
Roger, Jean-Michel, Boulet, Jean-Claude
openaire +1 more source
Short-Range Clutter Suppression for Airborne Radar Using Sparse Recovery and Orthogonal Projection
IEEE Geoscience and Remote Sensing Letters, 2022For nonside-looking airborne radar (NSLAR), the range-dependent near-range clutter degrades the performance of space–time adaptive processing (STAP), especially in the high-pulse-repetition-frequency mode.
Wei Chen, W. Xie, Yongliang Wang
semanticscholar +1 more source
An alternative paradigm of fault diagnosis in dynamic systems: orthogonal projection-based methods
at - Automatisierungstechnik, 2022In this paper, we propose a new paradigm of fault diagnosis in dynamic systems as an alternative to the well-established observer-based framework.
S. Ding, Linlin Li, Tianyu Liu
semanticscholar +1 more source
GopGAN: Gradients Orthogonal Projection Generative Adversarial Network With Continual Learning
IEEE Transactions on Neural Networks and Learning Systems, 2021The generative adversarial networks (GANs) in continual learning suffer from catastrophic forgetting. In continual learning, GANs tend to forget about previous generation tasks and only remember the tasks they just learned.
Xiaobin Li, Weiqiang Wang
semanticscholar +1 more source
IEEE Transactions on Neural Networks and Learning Systems, 2020
For the existing repetitive motion generation (RMG) schemes for kinematic control of redundant manipulators, the position error always exists and fluctuates.
Zhengtai Xie +4 more
semanticscholar +1 more source
For the existing repetitive motion generation (RMG) schemes for kinematic control of redundant manipulators, the position error always exists and fluctuates.
Zhengtai Xie +4 more
semanticscholar +1 more source
Certain Properties of Orthogonal Projections
Bulletin of the Iranian Mathematical Society, 2020Let \(P\), \(Q\) be projections on a Hilbert space. The pair \((P,Q)\) is Fredholm if the operator \(PQ|_{R(Q)}:R(Q)\to R(P)\) is Fredholm (here \(R(P)\) denotes the range of \(P\)). For a fixed projection \(P\), the authors discuss properties of the set of projections \(Q\) such that (a) \(P-Q\) is compact, (b) \((P,Q)\) is Fredholm, and some similar ...
Shuaijie Wang, Chunyuan Deng
openaire +2 more sources
Orthogonal Projection-Based Channel Estimation for Multi-Panel Millimeter Wave MIMO
IEEE Transactions on Communications, 2020Multi-panel MIMO is a promising technology in millimeter wave communications. Due to its partially hybrid structure and non-uniform antenna array, existing channel estimation cannot be directly applied to multi-panel MIMO. In this paper, we study channel
Wei Wang, Wei Zhang
semanticscholar +1 more source
Fault diagnosis of rotor based on Local-Global Balanced Orthogonal Discriminant Projection
, 2021The rotor is the most important part of the whole rotating machinery. Whether the rotor is normal directly determines the normal operation of the whole rotating machinery.
Mingkuan Shi +3 more
semanticscholar +1 more source
AAAI Conference on Artificial Intelligence
Low-Rank Adaptation (LoRA) enables efficient fine-tuning of large language models but suffers from catastrophic forgetting when learned updates interfere with the dominant singular directions that encode essential pre-trained knowledge.
Yifeng Xiong, Xiaohui Xie
semanticscholar +1 more source
Low-Rank Adaptation (LoRA) enables efficient fine-tuning of large language models but suffers from catastrophic forgetting when learned updates interfere with the dominant singular directions that encode essential pre-trained knowledge.
Yifeng Xiong, Xiaohui Xie
semanticscholar +1 more source