检索结果-内蒙古大学图书馆

Mentored learning: improving generalization and convergence of student learner via teaching feedback

学校读者我要写书评

暂无评论

The Journal of Machine Learning Research

The Journal of Machine Learning Research 2024年第1期25卷 15891-15935页

作者： Xiaofeng Cao Yaming Guo Heng Tao Shen Ivor W. Tsang James T. Kwok School of Artificial Intelligence Jilin University Changchun China and Australian Artifcial Intelligence Institute (AAII) University of Technology Sydney (UTS) NSW Australia School of Artificial Intelligence Jilin University Changchun China School of Computer Science and Technology Tongji University Shanghai China and School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China Centre for Frontier AI Research and Institute of High Performance Computing Agency for Science Technology and Research(A*STAR) Singapore Singapore and College of Computing and Data Science Nanyang Technological University Singapore Singapore Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong SAR China

Student learners typically engage in an iterative process of actively updating its hypotheses, like active learning. While this behavior can be advantageous, there is an inherent risk of introducing mistakes through incremental updates including weak initialization, inaccurate or insignificant history states, resulting in expensive convergence cost. In this work, rather than solely monitoring the update of the learner's status, we propose monitoring the disagreement w.r.t. $\mathcal{F}^\mathcal{T}(\cdot)$ between the learner and teacher, and call this new paradigm "Mentored Learning", which consists of 'how to teach' and 'how to learn'. By actively incorporating feedback that deviates from the learner's current hypotheses, convergence will be much easier to analyze without strict assumptions on learner's historical status, then deriving tighter generalization bounds on error and label complexity. Formally, we introduce an approximately optimal teaching hypothesis, hT, incorporating a tighter slack term $\left(1+\mathcal{F}^{\mathcal{T}}(\widehat{h}_t)\right)\Delta_t$ to replace the typical 2Δt used in hypothesis pruning. Theoretically, we demonstrate that, guided by this teaching hypothesis, the learner can converge to tighter generalization bounds on error and label complexity compared to non-educated learners who lack guidance from a teacher: 1) the generalization error upper bound can be reduced from R(h*)+4ΔT-1 to approximately R(hT )+2ΔT-1, and 2) the label complexity upper bound can be decreased from 4θ(TR(h*) + 2O(√T)) to approximately 2θ(2TR(hT) + 3O(√T)). To adhere strictly to our assumption, self-improvement of teaching is proposed when hT loosely approximates h*. In the context of learning, we further consider two teaching scenarios: instructing a white-box and black-box learner. Experiments validate this teaching concept and demonstrate superior generalization performance compared to fundamental active learning strategies, such as IWAL (Beygelzimer et al

关键词： machine teaching hypothesis pruning active learning error disagreement convergence generalization error label complexity

Causally-Aware Unsupervised Feature Selection Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Shen, Zongxin Huang, Yanyong Wang, Dongjie Ma, Minbo Lv, Fengmao Li, Tianrui Joint Laboratory of Data Science and Business Intelligence School of Statistics Southwestern University of Finance and Economics Chengdu611130 China Department of Electrical Engineering and Computer Science University of Kansas LawrenceKS66045 United States School of Computing and Artificial Intelligence Southwest Jiaotong University Chengdu611756 China

Unsupervised feature selection (UFS) has recently gained attention for its effectiveness in processing unlabeled high-dimensional data. However, existing methods overlook the intrinsic causal mechanisms within the data, resulting in the selection of irrelevant features and poor interpretability. Additionally, previous graph-based methods fail to account for the differing impacts of non-causal and causal features in constructing the similarity graph, which leads to false links in the generated graph. To address these issues, a novel UFS method, called Causally-Aware UnSupErvised Feature Selection learning (CAUSE-FS), is proposed. CAUSE-FS introduces a novel causal regularizer that reweights samples to balance the confounding distribution of each treatment feature. This regularizer is subsequently integrated into a generalized unsupervised spectral regression model to mitigate spurious associations between features and clustering labels, thus achieving causal feature selection. Furthermore, CAUSE-FS employs causality-guided hierarchical clustering to partition features with varying causal contributions into multiple granularities. By integrating similarity graphs learned adaptively at different granularities, CAUSE-FS increases the importance of causal features when constructing the fused similarity graph to capture the reliable local structure of data. Extensive experimental results demonstrate the superiority of CAUSE-FS over state-of-the-art methods, with its interpretability further validated through feature visualization. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

A Novel Grey Multivariate Convolution Model Based on the Improved Marine Predators Algorithm for Predicting Fossil Co2 Emissions in China

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Cang, Haoze Zeng, Xiangyan Yan, Shuli School of Mathematics and Computing Science Guangxi Colleges and Universities Key Laboratory of Data Analysis and Computation Guilin University of Electronic Technology Guilin541004 China School of Management Science and Engineering Nanjing University of Information Science and Technology Nanjing210044 China

The effective prediction of fossil CO2 emissions provides a foundation for environmental protection authorities to make relevant management and decisions. CO2 emissions are influenced by various factors and exhibit a nonlinear and complex trend. First, the weighted conformable fractional accumulation (WCFA) method is proposed. The weight of old and new information can be regulated by adjusting the parameters. A nonlinear discrete grey Bernoulli multivariate convolution model is introduced to model the sequences processed by WCFA. Then, five parameters in the model need to be optimized. An improved marine predators algorithm is proposed through the tent map and an improved control parameter. The feasibility and predictive effectiveness of the model is demonstrated by comparing it with four improved grey multivariate convolution models. Finally, the total fossil CO2 emissions in China from 2022 to 2026 are predicted and analyzed. The results indicate a gradual decrease in fossil CO2 emissions over the next five years. Some reasonable suggestions are made based on the prediction results and sources of fossil CO2 emissions. © 2023, The Authors. All rights reserved.

关键词： Forecasting

A Hybrid Framework for Spatial Interpolation: Merging data-driven with Domain Knowledge

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Cong Du, Shuyi Song, Hongqing Wang, Yuhe National & Local Joint Engineering Lab for Big Data Analysis and Computing Technology Beijing100190 China School of Civil and Resource Engineering University of Science and Technology Beijing Beijing100083 China Institute for Scientific Computation Texas A&M University College StationTX77843 United States

Estimating spatially distributed information through the interpolation of scattered observation datasets often overlooks the critical role of domain knowledge in understanding spatial dependencies. Additionally, the features of these datasets are typically limited to the spatial coordinates of the scattered observation locations. In this paper, we propose a hybrid framework that integrates data-driven spatial dependency feature extraction with rule-assisted spatial dependency function mapping to augment domain knowledge. We demonstrate the superior performance of our framework in two comparative application scenarios, highlighting its ability to capture more localized spatial features in the reconstructed distribution fields. Furthermore, we underscore its potential to enhance nonlinear estimation capabilities through the application of transformed fuzzy rules and to quantify the inherent uncertainties associated with the observation datasets. Our framework introduces an innovative approach to spatial information estimation by synergistically combining observational data with rule-assisted domain knowledge. © 2024, CC BY.

关键词： Domain Knowledge

HairDiffusion: Vivid Multi-Colored Hair Editing via Latent Diffusion

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zeng, Yu Zhang, Yang Liu, Jiachen Shen, Linlin Deng, Kaijun He, Weizhao Wang, Jinbao Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence and Robotics for Society China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China

Hair editing is a critical image synthesis task that aims to edit hair color and hairstyle using text descriptions or reference images, while preserving irrelevant attributes (e.g., identity, background, cloth). Many existing methods are based on StyleGAN to address this task. However, due to the limited spatial distribution of StyleGAN, it struggles with multiple hair color editing and facial preservation. Considering the advancements in diffusion models, we utilize Latent Diffusion Models (LDMs) for hairstyle editing. Our approach introduces Multi-stage Hairstyle Blend (MHB), effectively separating control of hair color and hairstyle in diffusion latent space. Additionally, we train a warping module to align the hair color with the target region. To further enhance multi-color hairstyle editing, we fine-tuned a CLIP model using a multi-color hairstyle dataset. Our method not only tackles the complexity of multi-color hairstyles but also addresses the challenge of preserving original colors during diffusion editing. Extensive experiments showcase the superiority of our method in editing multi-color hairstyles while preserving facial attributes given textual descriptions and reference images. © 2024, CC BY.

关键词： Color image processing

A Novel Effectiveness Assessment Framework for Neoadjuvant Chemoradiotherapy of Locally Advanced Rectal Cancer Based on Multi-modal Intelligence

学校读者我要写书评

暂无评论

A Novel Effectiveness Assessment Framework for Neoadjuvant C...

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Xiao Tian Dong Sui Weifeng Liu Maozu Guo Gongning Luo Kuanquan Wang Beijing Key Laboratory for Intelligent Processing Methods of Architectural Big Data Beijing China School of Electrical and Information Engineering Beijing University of Civil Engineering and Architecture Beijing China Perceptual Computing Research Center Harbin Institute of Technology Harbin China

ISBN: (数字)9798350337488

ISBN: (纸本)9798350337495

Neoadjuvant chemoradiotherapy (nCRT) is the stan-dard treatment for locally advanced rectal cancer (LARC). With the development of artificial intelligence, an increasing number of studies have begun to explore its application in cancer treatment prediction. However, the prior methods exhibit considerable variability even with slight modifications to the input data, which could potentially undermine the reliability of the results. In this paper, we proposed RP-Net, a novel multi-modal fusion-based framework that combines feature information from magnetic resonance imaging (MRI) and whole slide images (WSI), establishing a relationship to map the therapeutic effectiveness of nCRT for LARC. We investigated the relationship of the tumour region and its periphery tissues, and demonstrated the validity of the proposed framework that involving 11 different combinations of modalities. The experimental results revealed that it has achieved higher prediction accuracy compared to the four intra-categories single-modal combinations and outperformed the two intra-categories multi-modal combinations. When compared to the other four inter-categories multi-modal combinations, the fusion features get accuracy of 2 % ~ 6% improvement respectively.

关键词： Protocols Magnetic resonance imaging Decision making Cancer treatment Reliability Bioinformatics Artificial intelligence

Reliable Indoor Localization in Multi-Building Environments: Leveraging Environment-Invariant and Position-Related Features

学校读者我要写书评

暂无评论

IEEE Internet of Things Journal 2025年

作者： Long, Wenhan Wen, Xinlong Liu, Hao Li, Mengyao Li, Songquan Wu, Yonghui Chen, Fuxiang Liu, Lu Zhu, Rongbo Huazhong Agricultural University College of Informatics Wuhan430070 China Ministry of Education Engineering Research Center of Intelligent Technology for Agriculture Wuhan430070 China Hubei Engineering Technology Research Center of Agricultural Big Data Wuhan430070 China University of Leicester School of Computing and Mathematical Sciences Leicester United Kingdom University of Exeter Department of Computer Science Exeter United Kingdom

Received Signal Strength Indicator (RSSI)-based indoor localization offers a cost-effective solution for autonomous mobile robot navigation in 3D indoor environments, including cross-floor and multi-building structures. However, localization accuracy is fundamentally constrained by the low sampling density and unstable measurement of RSSI data. So far, existing methods neglect cross-environment RSSI coherence (e.g., repeated signal patterns in geometrically similar areas), resulting in unreliable fingerprint databases. What’s more, most approaches fail to model the spatial hierarchy of buildings, floors, and coordinates, which leads to lower accuracy in indoor positioning model predictions. To address these issues, we propose EP-3DLoc, a novel 3D indoor localization framework that combines an Environment-Invariant feature-based data Completion (EIC) method with a Position-Related feature-based Localization (PRL) method. The EIC enhances data quality by filling in sparse RSSI data using environment-invariant features, which are recurring RSSI patterns found in similar environmental structures. The PRL module combines multi-scale RSSI signal processing (raw data and image-like data) with a multi-task network that analyzes location relationships, enhancing localization accuracy in 3D environments. Experimental results on public datasets (TUT2018, UTSIndoorLoc, and UJIIndoorLoc) have demonstrated that EP-3DLoc achieves state-of-the-art performance on indoor localization in multi-building environments. Further testing on the self-constructed dataset HZAUIndoorLoc have revealed that EP-3DLoc not only outperforms existing methods in localization accuracy but also maintains low energy consumption and strong resistance to interference. © 2014 IEEE.

关键词： Floors

From General to Specific: Tailoring Large Language Models for Personalized Healthcare

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Shi, Ruize Huang, Hong Zhou, Wei Yin, Kehan Zhao, Kai Zhao, Yun Huazhong University of Science and Technology Wuhan China Tongji Medical College China Hubei Maternity and Child Health Care Hospital China National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology China

The rapid development of large language models (LLMs) has transformed many industries, including healthcare. However, previous medical LLMs have largely focused on leveraging general medical knowledge to provide responses, without accounting for patient variability and lacking true personalization at the individual level. To address this, we propose a novel method called personalized medical language model (PMLM), which explores and optimizes personalized LLMs through recommendation systems and reinforcement learning (RL). Specifically, by utilizing self-informed and peer-informed personalization, PMLM captures changes in behaviors and preferences to design initial personalized prompts tailored to individual needs. We further refine these initial personalized prompts through RL, ultimately enhancing the precision of LLM guidance. Notably, the personalized prompt are hard prompt, which grants PMLM high adaptability and reusability, allowing it to directly leverage high-quality proprietary LLMs. We evaluate PMLM using real-world obstetrics and gynecology data, and the experimental results demonstrate that PMLM achieves personalized responses, and it provides more refined and individualized services, offering a potential way for personalized medical LLMs. Copyright © 2024, The Authors. All rights reserved.

关键词： Reusability

Target Networks and Over-parameterization Stabilize Off-policy Bootstrapping with Function Approximation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Che, Fengdi Xiao, Chenjun Mei, Jincheng Dai, Bo Gummadi, Ramki Ramirez, Oscar A. Harris, Christopher K. Mahmood, A. Rupam Schuurmans, Dale Department of Computing Science University of Alberta United States School of Data Science The Chinese University of Hong Kong Shenzhen China Google DeepMind United Kingdom School of Computational Science and Engineering Georgia Tech United States Figure Uber United States CIFAR AI Chair Amii Canada

We prove that the combination of a target network and over-parameterized linear function approximation establishes a weaker convergence condition for bootstrapped value estimation in certain cases, even with off-policy data. Our condition is naturally satisfied for expected updates over the entire state-action space or learning with a batch of complete trajectories from episodic Markov decision processes. Notably, using only a target network or an over-parameterized model does not provide such a convergence guarantee. Additionally, we extend our results to learning with truncated trajectories, showing that convergence is achievable for all tasks with minor modifications, akin to value truncation for the final states in trajectories. Our primary result focuses on temporal difference estimation for prediction, providing high-probability value estimation error bounds and empirical analysis on Baird’s counterexample and a Four-room task. 1 Furthermore, we explore the control setting, demonstrating that similar convergence conditions apply to Q-learning. © 2024, CC BY-NC-ND.

关键词： Markov processes