检索结果-内蒙古大学图书馆

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Sili Huang Yanchao Sun Jifeng Hu Siyuan Guo Hechang Chen Yi Chang Lichao Sun Bo Yang Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education and School of Artificial Intelligence Jilin University China Department of Computer Science University of Maryland College Park School of Artificial Intelligence Jilin University China Lehigh University Bethlehem Pennsylvania Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education

In visual-based Reinforcement Learning (RL), agents often struggle to generalize well to environmental variations in the state space that were not observed during training. The variations can arise in both task-irrelevant features, such as background noise, and task-relevant features, such as robot configurations, that are related to the optimal decisions. To achieve generalization in both situations, agents are required to accurately understand the impact of changed features on the decisions, i.e., establishing the true associations between changed features and decisions in the policy model. However, due to the inherent correlations among features in the state space, the associations between features and decisions become entangled, making it difficult for the policy to distinguish them. To this end, we propose Saliency-Guided Features Decorrelation (SGFD) to eliminate these correlations through sample reweighting. Concretely, SGFD consists of two core techniques: Random Fourier Functions (RFF) and the saliency map. RFF is utilized to estimate the complex non-linear correlations in high-dimensional images, while the saliency map is designed to identify the changed features. Under the guidance of the saliency map, SGFD employs sample reweighting to minimize the estimated correlations related to changed features, thereby achieving decorrelation in visual RL tasks. Our experimental results demonstrate that SGFD can generalize well on a wide range of test environments and significantly outperforms state-of-the-art methods in handling both task-irrelevant variations and task-relevant variations.

关键词：

来源：评论

学校读者我要写书评

暂无评论

BA-LORA: BIAS-ALLEVIATING LOW-RANK ADAPTATION TO MITIGATE CATASTROPHIC INHERITANCE IN LARGE LANGUAGE MODELS

arXiv

引用

arXiv 2024年

作者： Chang, Yupeng Chang, Yi Wu, Yuan School of Artificial Intelligence Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering Jilin University China International Center of Future Science Jilin University China

Large language models (LLMs) have demonstrated remarkable proficiency across various natural language processing (NLP) tasks. However, adapting LLMs to downstream applications requires computationally intensive and memory-demanding fine-tuning procedures. To alleviate these burdens, parameter-efficient fine-tuning (PEFT) techniques have emerged as a promising approach to tailor LLMs with minimal computational overhead. While PEFT methods offer substantial advantages, they do not fully address the pervasive issue of bias propagation from pre-training data. This work introduces Bias-Alleviating Low-Rank Adaptation (BA-LoRA), a novel PEFT method designed to counteract bias inheritance. BA-LoRA incorporates three distinct regularization terms: (1) a consistency regularizer, (2) a diversity regularizer, and (3) a singular value decomposition regularizer. These regularizers aim to enhance the models’ consistency, diversity, and generalization capabilities during fine-tuning. We conduct extensive experiments on natural language understanding (NLU) and natural language generation (NLG) tasks using prominent LLMs such as LLaMA, Mistral, and Gemma. The results demonstrate that BA-LoRA outperforms LoRA and its state-of-the-art variants. Moreover, the extended experiments demonstrate that our method effectively mitigates the adverse effects of pre-training bias, leading to more reliable and robust model outputs. The code is available at https://***/cyp-jlu-ai/BA-LoRA. Copyright © 2024, The Authors. All rights reserved.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Large Language Model Evaluation via Matrix Nuclear-Norm

arXiv

引用

arXiv 2024年

作者： Li, Yahan Xia, Tingyu Chang, Yi Wu, Yuan School of Artificial Intelligence Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering Jilin University China International Center of Future Science Jilin University China

As large language models (LLMs) continue to evolve, efficient evaluation metrics are vital for assessing their ability to compress information and reduce redundancy. While traditional metrics like Matrix Entropy offer valuable insights, they are computationally intensive for large-scale models due to their O(n3) time complexity with Singular Value Decomposition (SVD). To mitigate this issue, we introduce the Matrix Nuclear-Norm, which not only serves as a metric to quantify the data compression proficiency of LLM but also provides a convex approximation of matrix rank to capture both predictive discriminability and diversity. By employing the L1,2-norm to further approximate the nuclear norm, we can effectively assess the model’s information compression capabilities. This approach reduces the time complexity to O(n2) and eliminates the need for SVD computation. Consequently, the Matrix Nuclear-Norm achieves speeds 8 to 24 times faster than Matrix Entropy for the CEREBRAS-GPT model as sizes increase from 111M to 6.7B. This performance gap becomes more pronounced with larger models, as validated in tests with other models like Pythia. Additionally, evaluations on benchmarks and model responses confirm that our proposed Matrix Nuclear-Norm is a reliable, scalable, and efficient tool for assessing LLMs’ performance, striking a balance between accuracy and computational efficiency. The code is available at https://***/MLGroupJLU/MatrixNuclearNorm. Copyright © 2024, The Authors. All rights reserved.

关键词： Singular value decomposition

来源：评论

学校读者我要写书评

暂无评论

From Incomplete Coarse-Grained to Complete Fine-Grained: A Two-Stage Framework for Spatiotemporal Data Reconstruction

arXiv

引用

arXiv 2024年

作者： Sun, Ziyu Su, Haoyang Wang, En Yang, Funing Yang, Yongjian Liu, Wenbin The College of Computer Science and Technology Jilin University Changchun130012 China The Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

With the rapid development of various sensing devices, spatiotemporal data is becoming increasingly important nowadays. However, due to sensing costs and privacy concerns, the collected data is often incomplete and coarse-grained, limiting its application to specific tasks. To address this, we propose a new task called spatiotemporal data reconstruction, which aims to infer complete and fine-grained data from sparse and coarse-grained observations. To achieve this, we introduce a two-stage data inference framework, DiffRecon, grounded in the Denoising Diffusion Probabilistic Model (DDPM). In the first stage, we present Diffusion-C, a diffusion model augmented by ST-PointFormer, a powerful encoder designed to leverage the spatial correlations between sparse data points. Following this, the second stage introduces Diffusion-F, which incorporates the proposed T-PatternNet to capture the temporal pattern within sequential data. Together, these two stages form an end-to-end framework capable of inferring complete, fine-grained data from incomplete and coarse-grained observations. We conducted experiments on multiple real-world datasets to demonstrate the superiority of our method. © 2024, CC BY.

关键词： Spatio-temporal data

来源：评论

学校读者我要写书评

暂无评论

Reusable Generator Data-Free knowledge Distillation with Hard Loss Simulation for Image Classification

SSRN

引用

SSRN 2024年

作者： Sun, Yafeng Wang, Xingwang Huang, Junhong Chen, Shilin Hou, Minghui College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

In many image classification scenarios where knowledge distillation (KD) is applied, multiple users need to train various student models that conform to the device's computational limitations at different times. However, due to security concerns, existing data-free KD (DFKD) methods discard the trained generator after completing a task, preventing its reuse in training subsequent students and resulting in wasted resources. Furthermore, the unlabeled nature of fake images eliminates the hard loss that plays a crucial role in data-driven KD. This paper introduces the reusable generator DFKD with hard loss simulation (RG-HLS) to address these challenges. RG-HLS employs a generator consisting of two generative networks and a decoder. The decoder and one of the generative networks can be reused in subsequent tasks. The adversarial example attack is also introduced to mimic the behavior of the teacher model in predicting real images, ensuring that DFKD exhibits a hard loss similar to that of data-driven KD. Extensive experiments across CIFAR-10, CIFAR-100, and Tiny-ImageNet datasets validate the proposed method's superior performance. The reusable generator accelerates convergence without compromising security, and integrating the hard loss further boosts accuracy. Code is available at https://***/sunyafeng-jlu/RG-HLS. © 2024, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

AWEQ: Post-Training Quantization with Activation-Weight Equalization for Large Language Models

arXiv

引用

arXiv 2023年

作者： Li, Baisong Wang, Xingwang Xu, Haixiao School of Computer Science and Technology Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University China

Large language models(LLMs) excellent performance across a variety of tasks, but they come with significant computational and storage costs. Quantizing these models is an effective way to alleviate this issue. However, existing methods struggle to strike a balance between model accuracy and hardware efficiency. This is where we introduce AWEQ, a post-training method that requires no additional training overhead. AWEQ excels in both ultra-low-bit quantization and 8-bit weight and activation (W8A8) quantization. There is an observation that weight quantization is less challenging than activation quantization. AWEQ transfers the difficulty of activation quantization to weights using channel equalization, achieving a balance between the quantization difficulties of both, and thereby maximizing performance. We have further refined the equalization method to mitigate quantization bias error, ensuring the robustness of the model. Extensive experiments on popular models such as LLaMA and OPT demonstrate that AWEQ outperforms all existing post-training quantization methods for large models. Copyright © 2023, The Authors. All rights reserved.

关键词： Chemical activation

来源：评论

学校读者我要写书评

暂无评论

Toward Time-Continuous Data Inference in Sparse Urban CrowdSensing

arXiv

引用

arXiv 2024年

作者： Sun, Ziyu Su, Haoyang Sun, Hanqi Wang, En Liu, Wenbin The College of Computer Science and Technology Jilin University Changchun130012 China The Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

Mobile Crowd Sensing (MCS) is a promising paradigm that leverages mobile users and their smart portable devices to perform various real-world tasks. However, due to budget constraints and the inaccessibility of certain areas, Sparse MCS has emerged as a more practical alternative, collecting data from a limited number of target subareas and utilizing inference algorithms to complete the full sensing map. While existing approaches typically assume a time-discrete setting with data remaining constant within each sensing cycle, this simplification can introduce significant errors, especially when dealing with long cycles, as real-world sensing data often changes continuously. In this paper, we go from fine-grained completion, i.e., the subdivision of sensing cycles into minimal time units, towards a more accurate, time-continuous completion. We first introduce Deep Matrix Factorization (DMF) as a neural network-enabled framework and enhance it with a Recurrent Neural Network (RNN-DMF) to capture temporal correlations in these finer time slices. To further deal with the continuous data, we propose TIME-DMF, which captures temporal information across unequal intervals, enabling time-continuous completion. Additionally, we present the Query-Generate (Q-G) strategy within TIME-DMF to model the infinite states of continuous data. Extensive experiments across five types of sensing tasks demonstrate the effectiveness of our models and the advantages of time-continuous completion. Copyright © 2024, The Authors. All rights reserved.

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

Learning Interpretable Network Dynamics via Universal Neural symbolic Regression

arXiv

引用

arXiv 2024年

作者： Hu, Jiao Cui, Jiaxu Yang, Bo College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

Discovering governing equations of complex network dynamics is a fundamental challenge in contemporary science with rich data, which can uncover the mysterious patterns and mechanisms of the formation and evolution of complex phenomena in various fields and assist in decision-making. In this work, we develop a universal computational tool that can automatically, efficiently, and accurately learn the symbolic changing patterns of complex system states by combining the excellent fitting ability from deep learning and the equation inference ability from pre-trained symbolic regression. We conduct intensive experimental verifications on more than ten representative scenarios from physics, biochemistry, ecology, epidemiology, etc. Results demonstrate the outstanding effectiveness and efficiency of our tool by comparing with the state-of-the-art symbolic regression techniques for network dynamics. The application to real-world systems including global epidemic transmission and pedestrian movements has verified its practical applicability. We believe that our tool can serve as a universal solution to dispel the fog of hidden mechanisms of changes in complex phenomena, advance toward interpretability, and inspire more scientific discoveries. Copyright © 2024, The Authors. All rights reserved.

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

AHMSA-Net: Adaptive Hierarchical Multi-Scale Attention Network for Micro-Expression Recognition

arXiv

引用

arXiv 2025年

作者： Zhang, Lijun Zhang, Yifan Tang, Weicheng Sun, Xinzhi Wang, Xiaomeng Li, Zhanshan College of Computer Science and Technology Jilin University Changchun Jilin130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun Jilin130012 China

Micro-expression recognition (MER) presents a significant challenge due to the transient and subtle nature of the motion changes involved. In recent years, deep learning methods based on attention mechanisms have made some breakthroughs in MER. However, these methods still suffer from the limitations of insufficient feature capture and poor dynamic adaptation when coping with the instantaneous subtle movement changes of micro-expressions. Therefore, in this paper, we design an Adaptive Hierarchical Multi-Scale Attention Network (AHMSA-Net) for MER. Specifically, we first utilize the onset and apex frames of the micro-expression sequence to extract threedimensional (3D) optical flow maps, including horizontal optical flow, vertical optical flow, and optical flow strain. Subsequently, the optical flow feature maps are inputted into AHMSA-Net, which consists of two parts: an adaptive hierarchical framework and a multi-scale attention mechanism. Based on the adaptive downsampling hierarchical attention framework, AHMSA-Net captures the subtle changes of micro-expressions from different granularities (fine and coarse) by dynamically adjusting the size of the optical flow feature map at each layer. Based on the multi-scale attention mechanism, AHMSA-Net learns micro-expression action information by fusing features from different scales (channel and spatial). These two modules work together to comprehensively improve the accuracy of MER. Additionally, rigorous experiments demonstrate that the proposed method achieves competitive results on major micro-expression databases, with AHMSA-Net achieving recognition accuracy of up to 78.21% on composite databases (SMIC, SAMM, CASMEII) and 77.08% on the CASME3 database. Copyright © 2025, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing Unsupervised Graph Few-shot Learning via Set Functions and Optimal Transport

arXiv

引用

arXiv 2025年

作者： Liu, Yonghao Giunchiglia, Fausto Li, Ximing Huang, Lan Feng, Xiaoyue Guan, Renchu College of Computer Science and Technology Jilin University Changchun China Department of Information Engineering and Computer Science University of Trento Trento Italy Key Laboratory of Symbolic Computation and Knowledge Engineering The Ministry of Education China

Graph few-shot learning has garnered significant attention for its ability to rapidly adapt to downstream tasks with limited labeled data, sparking considerable interest among researchers. Recent advancements in graph few-shot learning models have exhibited superior performance across diverse applications. Despite their successes, several limitations still exist. First, existing models in the meta-training phase predominantly focus on instance-level features within tasks, neglecting crucial set-level features essential for distinguishing between different categories. Second, these models often utilize query sets directly on classifiers trained with support sets containing only a few labeled examples, overlooking potential distribution shifts between these sets and leading to suboptimal performance. Finally, previous models typically require necessitate abundant labeled data from base classes to extract transferable knowledge, which is typically infeasible in real-world scenarios. To address these issues, we propose a novel model named STAR, which leverages Set funcTions and optimAl tRansport for enhancing unsupervised graph few-shot learning. Specifically, STAR utilizes expressive set functions to obtain set-level features in an unsupervised manner and employs optimal transport principles to align the distributions of support and query sets, thereby mitigating distribution shift effects. Theoretical analysis demonstrates that STAR can capture more task-relevant information and enhance generalization capabilities. Empirically, extensive experiments across multiple datasets validate the effectiveness of STAR. Our code can be found here. Copyright © 2025, The Authors. All rights reserved.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：