检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Joo, Sunghwan Jeong, Seokhyeon Heo, Juyeon Weller, Adrian Moon, Taesup Department of Electrical and Computer Engineering Sungkyunkwan University Korea Republic of Department of Electrical and Computer Engineering ASRI INMC IPAI Seoul National University Korea Republic of University of Cambridge United Kingdom The Alan Turing Institute United Kingdom

Neural network interpretation methods, particularly feature attribution methods, are known to be fragile with respect to adversarial input perturbations. To address this, several methods for enhancing the local smoothness of the gradient while training have been proposed for attaining robust feature attributions. However, the lack of considering the normalization of the attributions, which is essential in their visualizations, has been an obstacle to understanding and improving the robustness of feature attribution methods. In this paper, we provide new insights by taking such normalization into account. First, we show that for every non-negative homogeneous neural network, a naive 2-robust criterion for gradients is not normalization invariant, which means that two functions with the same normalized gradient can have different values. Second, we formulate a normalization invariant cosine distance-based criterion and derive its upper bound, which gives insight for why simply minimizing the Hessian norm at the input, as has been done in previous work, is not sufficient for attaining robust feature attribution. Finally, we propose to combine both 2 and cosine distance-based criteria as regularization terms to leverage the advantages of both in aligning the local gradient. As a result, we experimentally show that models trained with our method produce much more robust interpretations on CIFAR-10 and ImageNet-100 without significantly hurting the accuracy, compared to the recent baselines. To the best of our knowledge, this is the first work to verify the robustness of interpretation on a larger-scale dataset beyond CIFAR-10, thanks to the computational efficiency of our method. © 2022, CC BY-NC-SA.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Stability of linear systems with slow and fast time variation and switching

Stability of linear systems with slow and fast time variatio...

引用

IEEE Conference on Decision and Control

作者： Daniel Liberzon Hyungbo Shim Coordinated Science Laboratory University of Illinois Urbana-Champaign Urbana USA Department of Electrical and Computer Engineering ASRI Seoul National University Korea

ISBN: (数字)9781665467612

ISBN: (纸本)9781665467629

We establish exponential stability for a class of linear systems with slow and fast time variation and switching. We use the averaging method to approximate the original system by the average system which only exhibits slow time variation and switching. We then apply a stability criterion recently developed for such systems to prove stability of the average system and, consequently, of the original system.

关键词： Linear systems Stability Stability criteria Switches Control systems Control theory Nonlinear systems

来源：评论

学校读者我要写书评

暂无评论

Rethinking Training for De-biasing Text-to-Image Generation: Unlocking the Potential of Stable Diffusion

arXiv

引用

arXiv 2024年

作者： Kim, Eunji Kim, Siwon Park, Minjun Entezari, Rahim Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Stability AI United Kingdom AIIS ASRI INMC ISRC Seoul National University Korea Republic of

Recent advancements in text-to-image models, such as Stable Diffusion, show significant demographic biases. Existing debiasing techniques rely heavily on additional training, which imposes high computational costs and risks of compromising core image generation functionality. This hinders them from being widely adopted to real-world applications. In this paper, we explore Stable Diffusion’s overlooked potential to reduce bias without requiring additional training. Through our analysis, we uncover that initial noises associated with minority attributes form "minority regions" rather than scattered. We view these "minority regions" as opportunities in SD to reduce bias. To unlock the potential, we propose a novel de-biasing method called ‘weak guidance,’ carefully designed to guide a random noise to the minority regions without compromising semantic integrity. Through analysis and experiments on various versions of SD, we demonstrate that our proposed approach effectively reduces bias without additional training, achieving both efficiency and preservation of core image generation functionality. © 2024, CC BY-NC-SA.

关键词： Digital elevation model

来源：评论

学校读者我要写书评

暂无评论

ENTROPY IS NOT ENOUGH FOR TEST-TIME ADAPTATION: FROM THE PERSPECTIVE OF DISENTANGLED FACTORS

arXiv

引用

arXiv 2024年

作者： Lee, Jonghyun Jung, Dahuin Lee, Saehyung Park, Junsung Shin, Juhyeon Hwang, Uiwon Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of School of Computer Science and Engineering Soongsil University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Division of Digital Healthcare Yonsei University Korea Republic of AIIS ASRI INMC and ISRC Seoul National University Korea Republic of

Test-time adaptation (TTA) fine-tunes pre-trained deep neural networks for unseen test data. The primary challenge of TTA is limited access to the entire test dataset during online updates, causing error accumulation. To mitigate it, TTA methods have utilized the model output’s entropy as a confidence metric that aims to determine which samples have a lower likelihood of causing error. Through experimental studies, however, we observed the unreliability of entropy as a confidence metric for TTA under biased scenarios and theoretically revealed that it stems from the neglect of the influence of latent disentangled factors of data on predictions. Building upon these findings, we introduce a novel TTA method named Destroy Your Object (DeYO), which leverages a newly proposed confidence metric named Pseudo-Label Probability Difference (PLPD). PLPD quantifies the influence of the shape of an object on prediction by measuring the difference between predictions before and after applying an object-destructive transformation. DeYO consists of sample selection and sample weighting, which employ entropy and PLPD concurrently. For robust adaptation, DeYO prioritizes samples that dominantly incorporate shape information when making predictions. Our extensive experiments demonstrate the consistent superiority of DeYO over baseline methods across various scenarios, including biased and wild. Project page is publicly available at https://***/DeYO/. Copyright © 2024, The Authors. All rights reserved.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

Multi-Level Branched Regularization for Federated Learning

arXiv

引用

arXiv 2022年

作者： Kim, Jinkyu Kim, Geeho Han, Bohyung Computer Vision Laboratory Department of Electrical and Computer Engineering & ASRI Seoul National University Korea Republic of Interdisciplinary Program of Artificial Intelligence Seoul National University Korea Republic of

A critical challenge of federated learning is data heterogeneity and imbalance across clients, which leads to inconsistency between local networks and unstable convergence of global models. To alleviate the limitations, we propose a novel architectural regularization technique that constructs multiple auxiliary branches in each local model by grafting local and global subnetworks at several different levels and that learns the representations of the main pathway in the local model congruent to the auxiliary hybrid pathways via online knowledge distillation. The proposed technique is effective to robustify the global model even in the non-iid setting and is applicable to various federated learning frameworks conveniently without incurring extra communication costs. We perform comprehensive empirical studies and demonstrate remarkable performance gains in terms of accuracy and efficiency compared to existing methods. The source code is available in our project page1 Copyright © 2022, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Semi-Supervised Imitation Learning with Mixed Qualities of Demonstrations for Autonomous Driving

Semi-Supervised Imitation Learning with Mixed Qualities of D...

引用

International Conference on Control, Automation and Systems ( ICCAS)

作者： Gunmin Lee Wooseok Oh Jeongwoo Oh Seungyoun Shin Dohyeong Kim Jaeyeon Jeong Sungjoon Choi Songhwai Oh Department of Electrical and Computer Engineering and ASRI Seoul National University Seoul Korea School of Computer Science and Engineering Dongguk University Seoul Korea Department of Artificial Intelligence Korea University Seoul Korea

In this paper, we consider the problem of autonomous driving using imitation learning in a semi-supervised manner. In particular, both labeled and unlabeled demonstrations are leveraged during training by estimating the quality of each unlabeled demonstration. If the provided demonstrations are corrupted and have a low signal-to-noise ratio, the performance of the imitation learning agent can be degraded significantly. To mitigate this problem, we propose a method called semi-supervised imitation learning (SSIL). SSIL first learns how to discriminate and evaluate each state-action pair’s reliability in unlabeled demonstrations by assigning higher reliability values to demonstrations similar to labeled expert demonstrations. This reliability value is called leverage. After this discrimination process, labeled and unlabeled demonstrations with estimated leverage values are utilized while training the policy in a semi-supervised manner. The experimental results demonstrate the validity of the proposed algorithm using unlabeled trajectories with mixed qualities. Moreover, the hardware experiments using an RC car are conducted to show that the proposed method can be applied to real-world applications.

关键词： Training Automation Control systems Hardware Trajectory Reliability Automobiles

来源：评论

学校读者我要写书评

暂无评论

Reset & Distill: A Recipe for Overcoming Negative Transfer in Continual Reinforcement Learning

arXiv

引用

arXiv 2024年

作者： Ahn, Hongjoon Hyeon, Jinu Oh, Youngmin Hwang, Bosun Moon, Taesup Department of Electrical and Computer Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Samsung Advanced Institute of Technology Korea Republic of ASRI INMC IPAI/AIIS Seoul National University Korea Republic of

We argue that one of the main obstacles for developing effective Continual Reinforcement Learning (CRL) algorithms is the negative transfer issue occurring when the new task to learn arrives. Through comprehensive experimental validation, we demonstrate that such issue frequently exists in CRL and cannot be effectively addressed by several recent work on mitigating plasticity loss of RL agents. To that end, we develop Reset & Distill (R&D), a simple yet highly effective method, to overcome the negative transfer problem in CRL. R&D combines a strategy of resetting the agent’s online actor and critic networks to learn a new task and an offline learning step for distilling the knowledge from the online actor and previous expert’s action probabilities. We carried out extensive experiments on long sequence of Meta-World tasks and show that our method consistently outperforms recent baselines, achieving significantly higher success rates across a range of tasks. Our findings highlight the importance of considering negative transfer in CRL and emphasize the need for robust strategies like R&D to mitigate its detrimental effects. © 2024, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

On Task-Relevant Loss Functions in Meta-Reinforcement Learning and Online LQR

arXiv

引用

arXiv 2023年

作者： Shin, Jaeuk Kim, Giho Lee, Howon Han, Joonho Yang, Insoon The Department of Electrical and Computer Engineering Automation and Systems Research Institute Seoul National University Seoul08826 Korea Republic of Interdisciplinary Program in Artificial Intelligence ASRI Seoul National University Seoul08826 Korea Republic of

Designing a competent meta-reinforcement learning (meta-RL) algorithm in terms of data usage remains a central challenge to be tackled for its successful real-world applications. In this paper, we propose a sample-efficient meta-RL algorithm that learns a model of the system or environment at hand in a task-directed manner. As opposed to the standard model-based approaches to meta-RL, our method exploits the value information in order to rapidly capture the decision-critical part of the environment. The key component of our method is the loss function for learning the task inference module and the system model that systematically couples the model discrepancy and the value estimate, thereby facilitating the learning of the policy and the task inference module with a significantly smaller amount of data compared to the existing meta-RL algorithms. The idea is also extended to a non-meta-RL setting, namely an online linear quadratic regulator (LQR) problem, where our method can be simplified to reveal the essence of the strategy. The proposed method is evaluated in high-dimensional robotic control and online LQR problems, empirically verifying its effectiveness in extracting information indispensable for solving the tasks from observations in a sample efficient manner. Copyright © 2023, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

iCaps: An Interpretable Classifier via Disentangled Capsule Networks 1

引用

16th European Conference on computer Vision, ECCV 2020

作者： Jung, Dahuin Lee, Jonghyun Yi, Jihun Yoon, Sungroh Electrical and Computer Engineering ASRI INMC and Institute of Engineering Research Seoul National University Seoul08826 Korea Republic of

ISBN: (数字)9783030585297

ISBN: (纸本)9783030585280

We propose an interpretable Capsule Network, iCaps, for image classification. A capsule is a group of neurons nested inside each layer, and the one in the last layer is called a class capsule, which is a vector whose norm indicates a predicted probability for the class. Using the class capsule, existing Capsule Networks already provide some level of interpretability. However, there are two limitations which degrade its interpretability: 1) the class capsule also includes classification-irrelevant information, and 2) entities represented by the class capsule overlap. In this work, we address these two limitations using a novel class-supervised disentanglement algorithm and an additional regularizer, respectively. Through quantitative and qualitative evaluations on three datasets, we demonstrate that the resulting classifier, iCaps, provides a prediction along with clear rationales behind it with no performance degradation. © 2020, Springer Nature Switzerland AG.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Renderable Street View Map-Based Localization: Leveraging 3D Gaussian Splatting for Street-Level Positioning

Renderable Street View Map-Based Localization: Leveraging 3D...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Howoong Jun Hyeonwoo Yu Songhwai Oh Artificial Intelligence (IPAI) Seoul National University and Automation and Systems Research Institute (ASRI) and Sequor Robotics Inc. Seoul Korea Republic of Department of Intelligent Robotics & Mechanical Engineering Sungkyunkwan University (SKKU) Suwon-si Korea Republic of Department of Electrical and Computer Engineering (ECE) & Interdisciplinary Program in Artificial Intelligence (IPAI) Seoul National University and Automation and Systems Research Institute (ASRI) and Sequor Robotics Inc. Seoul Korea Republic of

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

In this paper, we introduce a new method that first utilizes 3D Gaussian splatting in street-level localization problem. Robust localization with street-level real-world images such as street view is a major issue for autonomous vehicle, augmented reality (AR) navigation, and outdoor mobile robots. The objective is to determine the position and orientation of a query image that matches a street view database composed of RGB images. However, given the limited information available in the street view images, accurately determining the location solely based on this data presents a significant challenge. To address this challenge, we propose a novel method called renderable street view map-based localization (RSM-Loc). This approach enhances the localization process by augmenting 2D street view images into a renderable 3D map using 3D Gaussian splatting, to resolve street-level localization problems. Upon receiving a query RGB image without geometry information, the proposed method renders 2D images from a pre-made renderable map and compares image pose similarities between the rendered images and the query image. Through iterations of this process, the proposed method eventually estimates the pose of the given query image. The experimental results demonstrate that RSM-Loc outperforms the baselines with neural-field-based localization. Additionally, we conduct deep analysis on the proposed method to show that our method can serve as a new concept for the street-level localization problem.

关键词： Location awareness Visualization Matched filters Three-dimensional displays Image resolution Navigation Noise measurement Mobile robots Intelligent robots Image reconstruction

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：