Výsledky vyhledávání

Report

LoopSR: Looping Sim-and-Real for Lifelong Policy Adaptation of Legged Robots

Autor: Wu, Peilin, Xie, Weiji, Cao, Jiahang, Lai, Hang, Zhang, Weinan

Reinforcement Learning (RL) has shown its remarkable and generalizable capability in legged locomotion through sim-to-real transfer. However, while adaptive methods like domain randomization are expected to make policy more robust to diverse environm

Externí odkaz: http://arxiv.org/abs/2409.17992

Zobrazit plný text záznamu

Report

IDEA: Enhancing the Rule Learning Ability of Large Language Model Agent through Induction, Deduction, and Abduction

Autor: He, Kaiyu, Zhang, Mian, Yan, Shuo, Wu, Peilin, Chen, Zhiyu Zoey

While large language models (LLMs) have been thoroughly evaluated for deductive and inductive reasoning, their proficiency in abductive reasoning and holistic rule learning in interactive environments remains less explored. We introduce RULEARN, a no

Externí odkaz: http://arxiv.org/abs/2408.10455

Zobrazit plný text záznamu

Report

Quantifying the Blockchain Trilemma: A Comparative Analysis of Algorand, Ethereum 2.0, and Beyond

Autor: Fu, Yihang, Jing, Mingwei, Zhou, Jiaolun, Wu, Peilin, Wang, Ye, Zhang, Luyao, Hu, Chuang

Blockchain technology is essential for the digital economy and metaverse, supporting applications from decentralized finance to virtual assets. However, its potential is constrained by the "Blockchain Trilemma," which necessitates balancing decentral

Externí odkaz: http://arxiv.org/abs/2407.14335

Zobrazit plný text záznamu

Report

Aligning Speakers: Evaluating and Visualizing Text-based Diarization Using Efficient Multiple Sequence Alignment (Extended Version)

Autor: Gong, Chen, Wu, Peilin, Choi, Jinho D.

This paper presents a novel evaluation approach to text-based speaker diarization (SD), tackling the limitations of traditional metrics that do not account for any contextual information in text. Two new metrics are proposed, Text-based Diarization E

Externí odkaz: http://arxiv.org/abs/2309.07677

Zobrazit plný text záznamu

Report

Bridging the Sim-to-Real Gap from the Information Bottleneck Perspective

Autor: He, Haoran, Wu, Peilin, Bai, Chenjia, Lai, Hang, Wang, Lingxiao, Pan, Ling, Hu, Xiaolin, Zhang, Weinan

Reinforcement Learning (RL) has recently achieved remarkable success in robotic control. However, most works in RL operate in simulated environments where privileged knowledge (e.g., dynamics, surroundings, terrains) is readily available. Conversely,

Externí odkaz: http://arxiv.org/abs/2305.18464

Zobrazit plný text záznamu