Zobrazeno 1 - 10
of 659
pro vyhledávání: '"Luo, Zhi-Quan"'
Foundation models often struggle with uncertainty when faced with new situations in online decision-making, necessitating scalable and efficient exploration to resolve this uncertainty. We introduce GPT-HyperAgent, an augmentation of GPT with HyperAg
Externí odkaz:
http://arxiv.org/abs/2407.13195
Conventional policy for configuring an intelligent reflecting surface (IRS) typically requires channel state information (CSI), thus incurring substantial overhead costs and facing incompatibility with the current network protocols. This paper propos
Externí odkaz:
http://arxiv.org/abs/2407.12648
Autor:
Zhang, Yushun, Chen, Congliang, Li, Ziniu, Ding, Tian, Wu, Chenwei, Ye, Yinyu, Luo, Zhi-Quan, Sun, Ruoyu
We propose Adam-mini, an optimizer that achieves on-par or better performance than AdamW with 45% to 50% less memory footprint. Adam-mini reduces memory by cutting down the learning rate resources in Adam (i.e., $1/\sqrt{v}$). We find that $\geq$ 90%
Externí odkaz:
http://arxiv.org/abs/2406.16793
In adversarial machine learning, neural networks suffer from a significant issue known as robust overfitting, where the robust test accuracy decreases over epochs (Rice et al., 2020). Recent research conducted by Xing et al.,2021; Xiao et al., 2022 h
Externí odkaz:
http://arxiv.org/abs/2405.01817
Autor:
Li, Yingru, Luo, Zhi-Quan
This work advances randomized exploration in reinforcement learning (RL) with function approximation modeled by linear mixture MDPs. We establish the first prior-dependent Bayesian regret bound for RL with function approximation; and refine the Bayes
Externí odkaz:
http://arxiv.org/abs/2403.11175
SGD performs worse than Adam by a significant margin on Transformers, but the reason remains unclear. In this work, we provide an explanation through the lens of Hessian: (i) Transformers are "heterogeneous": the Hessian spectrum across parameter blo
Externí odkaz:
http://arxiv.org/abs/2402.16788
The dynamic competition between radar and jammer systems presents a significant challenge for modern Electronic Warfare (EW), as current active learning approaches still lack sample efficiency and fail to exploit jammer's characteristics. In this pap
Externí odkaz:
http://arxiv.org/abs/2402.16274
This work tackles the complexities of multi-player scenarios in \emph{unknown games}, where the primary challenge lies in navigating the uncertainty of the environment through bandit feedback alongside strategic decision-making. We introduce Thompson
Externí odkaz:
http://arxiv.org/abs/2402.09456
We propose HyperAgent, a reinforcement learning (RL) algorithm based on the hypermodel framework for exploration in RL. HyperAgent allows for the efficient incremental approximation of posteriors associated with an optimal action-value function ($Q^\
Externí odkaz:
http://arxiv.org/abs/2402.10228
Autor:
Wu, Qingqing, Zheng, Beixiong, You, Changsheng, Zhu, Lipeng, Shen, Kaiming, Shao, Xiaodan, Mei, Weidong, Di, Boya, Zhang, Hongliang, Basar, Ertugrul, Song, Lingyang, Di Renzo, Marco, Luo, Zhi-Quan, Zhang, Rui
Intelligent surfaces (ISs) have emerged as a key technology to empower a wide range of appealing applications for wireless networks, due to their low cost, high energy efficiency, flexibility of deployment and capability of constructing favorable wir
Externí odkaz:
http://arxiv.org/abs/2312.16918