Zobrazeno 1 - 10
of 966
pro vyhledávání: '"Yu, XuDong"'
In Reinforcement Learning (RL), training a policy from scratch with online experiences can be inefficient because of the difficulties in exploration. Recently, offline RL provides a promising solution by giving an initialized offline policy, which ca
Externí odkaz:
http://arxiv.org/abs/2405.07223
Cross-domain offline reinforcement learning leverages source domain data with diverse transition dynamics to alleviate the data requirement for the target domain. However, simply merging the data of two domains leads to performance degradation due to
Externí odkaz:
http://arxiv.org/abs/2405.06192
Autor:
Yu, Xudong
In the Standard Model, charged lepton flavor violation (CLFV) is heavily suppressed by tiny neutrino mass, while many theoretical models can enhance CLFV effects up to a detectable level. The observation of any CLFV process would be a clear signal of
Externí odkaz:
http://arxiv.org/abs/2405.00272
Offline Reinforcement Learning (RL) faces distributional shift and unreliable value estimation, especially for out-of-distribution (OOD) actions. To address this, existing uncertainty-based methods penalize the value function with uncertainty quantif
Externí odkaz:
http://arxiv.org/abs/2404.06188
Sequential decision-making is desired to align with human intents and exhibit versatility across various tasks. Previous methods formulate it as a conditional generation process, utilizing return-conditioned diffusion models to directly model traject
Externí odkaz:
http://arxiv.org/abs/2404.04920
Autor:
Yu, Xudong, Wang, Zijian, Liu, Cheng-en, Feng, Yiqing, Li, Jinning, Geng, Xinyue, Zhang, Yimeng, Gao, Leyun, Jiang, Ruobing, Wu, Youpeng, Zhou, Chen, Li, Qite, Wang, Siguang, Ban, Yong, Mao, Yajun, Li, Qiang
We propose here a set of new methods to directly detect light mass dark matter through its scattering with abundant atmospheric muons or accelerator beams. Firstly, we plan to use the free cosmic-ray muons interacting with dark matter in a volume sur
Externí odkaz:
http://arxiv.org/abs/2402.13483
Publikováno v:
Journal of Artificial Intelligence Research, vol. 81, pp. 481-509, 2024
To obtain a near-optimal policy with fewer interactions in Reinforcement Learning (RL), a promising approach involves the combination of offline RL, which enhances sample efficiency by leveraging offline datasets, and online RL, which explores inform
Externí odkaz:
http://arxiv.org/abs/2309.16973
Autor:
Lecoutre, Baptiste, Guo, Yukun, Yu, Xudong, Niranjan, M., Mukhtar, Musawwadah, Volchkov, Valentin V., Aspect, Alain, Josse, Vincent
The ability to load ultracold atoms at a well-defined energy in a disordered potential is a crucial tool to study quantum transport, and in particular Anderson localization. In this paper, we present a new method for achieving that goal by rf transfe
Externí odkaz:
http://arxiv.org/abs/2208.12166
Realizing a large-scale fully controllable quantum system is a challenging task in current physical research and has broad applications. Ultracold atom and molecule arrays in optical tweezers in vacuum have been used for quantum simulation, quantum m
Externí odkaz:
http://arxiv.org/abs/2207.03641
We experimentally study the interference of dipole scattered light from two optically levitated nanoparticles in vacuum, which present an environment free of particle-substrate interactions. We illuminate the two trapped nanoparticles with a linearly
Externí odkaz:
http://arxiv.org/abs/2205.11348