检索结果-内蒙古大学图书馆

Contract-Inspired Contest Theory for Controllable Image Generation in Mobile Edge Metaverse

学校读者我要写书评

暂无评论

IEEE Transactions on Mobile Computing 2025年

作者： Liu, Guangyuan Du, Hongyang Wang, Jiacheng Niyato, Dusit Kim, Dong In Nanyang Technological University College of Computing and Data Science Energy Research Institute @ NTU Interdisciplinary Graduate Program Singapore University of Hong Kong Department of Electrical and Electronic Engineering Hong Kong Hong Kong Nanyang Technological University College of Computing and Data Science Singapore Sungkyunkwan University Department of Electrical and Computer Engineering Korea Republic of

The rapid advancement of immersive technologies has propelled the development of the Metaverse, where the convergence of virtual and physical realities necessitates the generation of high-quality, photorealistic images to enhance user experience. However, generating these images, especially through Generative Diffusion Models (GDMs), in mobile edge computing environments presents significant challenges due to the limited computing resources of edge devices and the dynamic nature of wireless networks. This paper proposes a novel framework that integrates contract-inspired contest theory, Deep Reinforcement Learning (DRL), and GDMs to optimize image generation in these resource-constrained environments. The framework addresses the critical challenges of resource allocation and semantic data transmission quality by incentivizing edge devices to efficiently transmit high-quality semantic data, which is essential for creating realistic and immersive images. The use of contest and contract theory ensures that edge devices are motivated to allocate resources effectively, while DRL dynamically adjusts to network conditions, optimizing the overall image generation process. Experimental results demonstrate that the proposed approach not only improves the quality of generated images but also achieves superior convergence speed and stability compared to traditional methods. This makes the framework particularly effective for optimizing complex resource allocation tasks in mobile edge Metaverse applications, offering enhanced performance and efficiency in creating immersive virtual environments. © 2002-2012 IEEE.

关键词： Resource allocation

Stage-Wise Reward Shaping for Acrobatic Robots: A Constrained Multi-Objective Reinforcement Learning Approach

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kim, Dohyeong Kwon, Hyeokjin Kim, Junseok Lee, Gunmin Oh, Songhwai The Department of Electrical and Computer Engineering and ASRI Seoul National University Seoul08826 Korea Republic of The Interdisciplinary Program in Artificial Intelligence and ASRI Seoul National University Seoul08826 Korea Republic of

As the complexity of tasks addressed through reinforcement learning (RL) increases, the definition of reward functions also has become highly complicated. We introduce an RL method aimed at simplifying the reward-shaping process through intuitive strategies. Initially, instead of a single reward function composed of various terms, we define multiple reward and cost functions within a constrained multi-objective RL (CMORL) framework. For tasks involving sequential complex movements, we segment the task into distinct stages and define multiple rewards and costs for each stage. Finally, we introduce a practical CMORL algorithm that maximizes objectives based on these rewards while satisfying constraints defined by the costs. The proposed method has been successfully demonstrated across a variety of acrobatic tasks in both simulation and real-world environments. Additionally, it has been shown to successfully perform tasks compared to existing RL and constrained RL algorithms. Our code is available at https://***/rllab-snu/Stage-Wise-CMORL. Copyright © 2024, The Authors. All rights reserved.

关键词： Cost functions

Few-Shot Open-Set Keyword Spotting with Multi-Stage Training

学校读者我要写书评

暂无评论

Few-Shot Open-Set Keyword Spotting with Multi-Stage Training

Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA)

作者： Lo-Ya Li Tien-Hong Lo Jeih-Weih Hung Shih-Chieh Huang Berlin Chen Institute of AI Interdisciplinary Applied Technology National Taiwan Normal University Taiwan Department of Computer Science and Information Engineering National Taiwan Normal University Taiwan Department of Electrical Engineering National Chi Nan University Taiwan Realtek Semiconductor Corp. Taiwan

ISBN: (数字)9798350367331

ISBN: (纸本)9798350367348

As the advance of human-computer interaction technologies continued, keyword spotting (KWS) systems have gained prominence in everyday devices. This study is dedicated to exploring innovative approaches for few-shot keyword recognition under open-set conditions, a challenging yet crucial area in speech processing. To this end, we design and develop a multi-stage training method that synergistically combines the advantages of acoustic and phonetic features, thereby substantially enhancing the ability of a KWS model. By learning multi-type features with joint training from only one dataset, our KWS model is equipped with a more robustness feature extractor to deal with few-shot KWS. Experimental results demonstrate that our model outperforms strong baselines by achieving a 15% improvement in recognition accuracy on open-set tests in a 10shot-10way setting. This research confirms the effectiveness of our multi-stage strategy and suggests promising directions for future development in keyword recognition technologies.

关键词： Training Accuracy Computational modeling Prototypes Phonetics Feature extraction Robustness Speech processing Few shot learning Testing

Interactive Facial Expression Editing with Non-linear Blendshape Interpolation

学校读者我要写书评

暂无评论

Interactive Facial Expression Editing with Non-linear Blends...

43rd Annual Conference on European Association for computer Graphics, EUROGRAPHICS 2022

作者： Roh, Ji Hyun Kim, Seong Uk Jang, Hanyoung Seol, Yeongho Kim, Jongmin Department of Computer Science and Engineering Kangwon National University Korea Republic of Interdisciplinary Graduate Program in Medical Bigdata Convergence Kangwon National University Korea Republic of Game AI Lab NC Korea Republic of NVIDIA Korea Republic of

ISBN: (纸本)9783038681694

The ability to manipulate facial animations interactively is vital for enhancing the productivity and quality of character animation. In this paper, we present a novel interactive facial animation editing system that can express the naturalness of non-linear facial movements in real-time. The proposed system is based on a fully automatic algorithm that maintains all positional constraints while deforming the facial mesh as realistic as possible. Our method is based on direct manipulation with non-linear blendshape interpolation. We formulate the facial animation editing as a two-step quadratic minimization and solve it efficiently. From our results, the proposed method produces the desired and realistic facial animation better compared to existing mesh deformation methods, which are mainly based on linear combination and optimization. © 2022 The Author(s).

关键词： Interpolation

Safe CoR: A Dual-Expert Approach to Integrating Imitation Learning and Safe Reinforcement Learning Using Constraint Rewards

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kwon, Hyeokjin Lee, Gunmin Lee, Junseo Oh, Songhwai Interdisciplinary Program in Artificial Intelligence ASRI Seoul National University Seoul08826 Korea Republic of Department of Electrical and Computer Engineering ASRI Seoul National University Seoul08826 Korea Republic of

In the realm of autonomous agents, ensuring safety and reliability in complex and dynamic environments remains a paramount challenge. Safe reinforcement learning addresses these concerns by introducing safety constraints, but still faces challenges in navigating intricate environments such as complex driving situations. To overcome these challenges, we present the safe constraint reward (Safe CoR) framework, a novel method that utilizes two types of expert demonstrations—reward expert demonstrations focusing on performance optimization and safe expert demonstrations prioritizing safety. By exploiting a constraint reward (CoR), our framework guides the agent to balance performance goals of reward sum with safety constraints. We test the proposed framework in diverse environments, including the safety gym, metadrive, and the real-world Jackal platform. Our proposed framework enhances the performance of algorithms by 39% and reduces constraint violations by 88% on the real-world Jackal platform, demonstrating the framework’s efficacy. Through this innovative approach, we expect significant advancements in real-world performance, leading to transformative effects in the realm of safe and reliable autonomous agents. Copyright © 2024, The Authors. All rights reserved.

关键词： Demonstrations

MathReader: Text-to-Speech for Mathematical Documents

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Hyeon, Sieun Jung, Kyudan Kim, Nam-Joon Ryu, Hyun Gon Do, Jaeyoung Department of Electrical and Computer Engineering Seoul National University Korea Republic of Department of Mathematics Chung-Ang University Korea Republic of NVIDIA United States Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of

TTS (Text-to-Speech) document reader from Microsoft, Adobe, Apple, and Openai have been serviced worldwide. They provide relatively good TTS results for general plain text, but sometimes skip contents or provide unsatisfactory results for mathematical expressions. This is because most modern academic papers are written in LaTeX, and when LaTeX formulas are compiled, they are rendered as distinctive text forms within the document. However, traditional TTS document readers output only the text as it is recognized, without considering the mathematical meaning of the formulas. To address this issue, we propose MathReader, which effectively integrates OCR, a fine-tuned T5 model, and TTS. MathReader demonstrated a lower Word Error Rate (WER) than existing TTS document readers, such as Microsoft Edge and Adobe Acrobat, when processing documents containing mathematical formulas. MathReader reduced the WER from 0.510 to 0.281 compared to Microsoft Edge, and from 0.617 to 0.281 compared to Adobe Acrobat. This will significantly contribute to alleviating the inconvenience faced by users who want to listen to documents, especially those who are visually impaired. The code is available at https://***/hyeonsieun/MathReader. Copyright © 2025, The Authors. All rights reserved.

关键词：

NEW INSIGHTS FOR THE STABILITY-PLASTICITY DILEMMA IN ONLINE CONTINUAL LEARNING

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jung, Dahuin Lee, Dongjin Hong, Sunwon Jang, Hyemi Bae, Ho Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of DeepBrain AI Inc. Department of Cyber Security Ewha Womans University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Seoul08826 Korea Republic of

The aim of continual learning is to learn new tasks continuously (i.e., plasticity) without forgetting previously learned knowledge from old tasks (i.e., stability). In the scenario of online continual learning, wherein data comes strictly in a streaming manner, the plasticity of online continual learning is more vulnerable than offline continual learning because the training signal that can be obtained from a single data point is limited. To overcome the stability-plasticity dilemma in online continual learning, we propose an online continual learning framework named multi-scale feature adaptation network (MuFAN) that utilizes a richer context encoding extracted from different levels of a pre-trained network. Additionally, we introduce a novel structure-wise distillation loss and replace the commonly used batch normalization layer with a newly proposed stability-plasticity normalization module to train MuFAN that simultaneously maintains high plasticity and stability. MuFAN outperforms other state-of-the-art continual learning methods on the SVHN, CIFAR100, miniImageNet, and CORe50 datasets. Extensive experiments and ablation studies validate the significance and scalability of each proposed component: 1) multi-scale feature maps from a pre-trained encoder, 2) the structure-wise distillation loss, and 3) the stability-plasticity normalization module in MuFAN. Code is publicly available at https://***/whitesnowdrop/MuFAN. Copyright © 2023, The Authors. All rights reserved.

关键词： Stability

ControlDreamer: Blending Geometry and Style in Text-to-3D

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Oh, Yeongtak Choi, Jooyoung Kim, Yongsung Park, Minjun Shin, Chaehun Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Seoul Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Seoul Korea Republic of

Recent advancements in text-to-3D generation have significantly contributed to the automation and democratization of 3D content creation. Building upon these developments, we aim to address the limitations of current methods in blending geometries and styles in text-to-3D generation. We introduce multi-view ControlNet, a novel depth-aware multi-view diffusion model trained on generated datasets from a carefully curated text corpus. Our multi-view ControlNet is then integrated into our two-stage pipeline, ControlDreamer, enabling text-guided generation of stylized 3D models. Additionally, we present a comprehensive benchmark for 3D style editing, encompassing a broad range of subjects, including objects, animals, and characters, to further facilitate research on diverse 3D generation. Our comparative analysis reveals that this new pipeline outperforms existing text-to-3D methods as evidenced by human evaluations and CLIP score metrics. Project page: https://***. © 2023, CC BY.

关键词： 3D modeling

Battling the Non-stationarity in Time Series Forecasting via Test-time Adaptation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Kim, HyunGi Kim, Siwon Mok, Jisoo 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 AIIS ASRI INMC Seoul National University Korea Republic of

Deep Neural Networks have spearheaded remarkable advancements in time series forecasting (TSF), one of the major tasks in time series modeling. Nonetheless, the non-stationarity of time series undermines the reliability of pre-trained source time series forecasters in mission-critical deployment settings. In this study, we introduce a pioneering test-time adaptation framework tailored for TSF (TSF-TTA). TAFAS, the proposed approach to TSF-TTA, flexibly adapts source forecasters to continuously shifting test distributions while preserving the core semantic information learned during pre-training. The novel utilization of partially-observed ground truth and gated calibration module enables proactive, robust, and model-agnostic adaptation of source forecasters. Experiments on diverse benchmark datasets and cutting-edge architectures demonstrate the efficacy and generality of TAFAS, especially in long-term forecasting scenarios that suffer from significant distribution shifts. The code is available at https://***/kimanki/TAFAS. © 2025, CC BY-NC-SA.

关键词： Deep neural networks