Bias to Balance: New-Knowledge-Preferred Few-Shot Class-Incremental Learning via Transition Calibration [0.03%]
从偏差到平衡：通过转换校准的新型知识优先的少量样本类渐进学习
Hongquan Zhang,Zhizhong Zhang,Xin Tan et al.
Hongquan Zhang et al.
Humans can quickly learn new concepts with limited experience, while not forgetting learned knowledge. Such ability in machine learning is referred to as few-shot class-incremental learning (FSCIL). Although some methods try to solve this p...
Data-Model Hybrid-Driven Safe Reinforcement Learning for Adaptive Avoidance Control Against Unsafe Moving Zones [0.03%]
基于数据与模型的混合驱动的安全强化学习在应对危险移动区域的自适应避碰控制中的应用研究
Ke Wang,Chaoxu Mu,Anguo Zhang et al.
Ke Wang et al.
With the gradual application of reinforcement learning (RL), safety has emerged as a paramount concern. This article presents a novel data-model hybrid-driven safe RL (SRL) scheme to address the challenge of avoidance control in the operati...
Xiangmin Han,Rundong Xue,Jingxi Feng et al.
Xiangmin Han et al.
The goal of the hypergraph foundation model (HGFM) is to learn an encoder based on the hypergraph computational paradigm through self-supervised pretraining on high-order correlation structures, enabling the encoder to rapidly adapt to vari...
Visual Reinforcement Learning Control With Instance-Reweighted Alignment and Instance-Dimension Uniformity [0.03%]
具有实例重权对齐和实例维度一致性的视觉强化学习控制
Rongrong Wang,Yuhu Cheng,Xuesong Wang
Rongrong Wang
Visual reinforcement learning (VRL) has demonstrated remarkable capabilities in learning behaviors directly from intricate high-dimensional visual inputs. Despite these advancements, existing VRL methods still encounter obstacles such as co...
Multithreaded Asynchronous Deep Reinforcement Learning With Multisensor Fusion for Robot Collision Avoidance [0.03%]
基于多传感器融合的多线程异步深度强化学习机器人避碰方法
Chao Sun,Xing Wu,Yanxu Su et al.
Chao Sun et al.
To develop a safe and efficient navigation system of robotic vehicles in dynamic scenes, a new collision-avoidance method using deep reinforcement learning (DRL) is presented. First, a novel method of DRL based on multithreaded asynchronous...
RefConv: Reparameterized Refocusing Convolution for Powerful ConvNets [0.03%]
Rep参数化 refocusing 卷积 for 强大的 卷积神经网络
Zhicheng Cai,Xiaohan Ding,Qiu Shen et al.
Zhicheng Cai et al.
We propose reparameterized refocusing convolution (RefConv) as a replacement for regular convolutional layers, which is a plug-and-play module to improve the performance without any inference costs. Specifically, given a pretrained model, R...
An Innovative Multisource Teacher Collaborative Framework for Self-Knowledge Distillation [0.03%]
一种创新的多源教师协同框架用于自知识提炼
Lei Zhao,Wing W Y Ng,Jianjun Zhang et al.
Lei Zhao et al.
Self-knowledge distillation, abbreviated as SKD, exhibits greater computational efficiency than traditional knowledge distillation (KD) because it learns from its own predictions rather than from a pretrained teacher. Existing SKD methods d...
S4DL: Shift-Sensitive Spatial-Spectral Disentangling Learning for Hyperspectral Image Unsupervised Domain Adaptation [0.03%]
基于移位敏感的超光谱图像无监督领域自适应空间-光谱解耦学习（S4DL）
Jie Feng,Tianshu Zhang,Junpeng Zhang et al.
Jie Feng et al.
Unsupervised domain adaptation (UDA) techniques, extensively studied in hyperspectral image (HSI) classification, aim to use labeled source domain data and unlabeled target domain data to learn domain invariant features for cross-scene clas...
Ugochukwu Ejike Akpudo,Yongsheng Gao,Jun Zhou et al.
Ugochukwu Ejike Akpudo et al.
Convolutional neural networks (CNNs) have succeeded remarkably in various computer vision tasks. However, they are not intrinsically explainable. While feature-level understanding of CNNs reveals where the models looked, concept-based expla...
Heterogeneous Mutual Knowledge Distillation for Wearable Human Activity Recognition [0.03%]
异构相互知识蒸馏的可穿戴人类活动识别方法
Zhiwen Xiao,Huanlai Xing,Rong Qu et al.
Zhiwen Xiao et al.
Recently, numerous deep learning algorithms have addressed wearable human activity recognition (HAR), but they often struggle with efficient knowledge transfer to lightweight models for mobile devices. Knowledge distillation (KD) is a popul...