Sihan Liu

Sihan Liu

Auto-generated profile, based on publicly available information

Sihan Liu

Department of Computer Science and Engineering UCSD 92092. CA

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 January 2019 in IEEE Access
Verified
Journal Article
Open access
SLSihan Liu
Sihan Liu
YWYucheng Wu
Yucheng Wu
LLLiang Li
Liang Li
XLXiaocui Liu
Xiaocui Liu
WXWeiyang Xu
Weiyang Xu
A large number of mobile multimedia terminals are prominent features of smart cities. Device-to-device (D2D) communication takes advantage of the limited bandwidth resources of cellular networks to accommodate more mobile devices. However, when D2D pairs reuse cellular users channels, serious interference leads to energy consumption, which dissatisfies the requirements of green communication. This study focuses on energy efficiency maximization of D2D communication under the constraints of both D2D pairs and cellular users quality of service (QoS). The formulated resource allocation problem is NP-hard, which is usually difficult to solve within polynomial time. To make the problem easy to handle, we divide it into power control and channel allocation sub-problems. In particular, we propose a power control algorithm based on LambertWfunction to maximize the energy efficiency of a single D2D pair. The preference values of D2D pairs and cellular users are calculated using the power control results, respectively. A channel allocation scheme based on Gale-Shapley algorithm utilizes preference values to match two sides, which aims at maximizing the signal to interference plus noise ratio (SINR) of cellular users and the energy efficiency of D2D pairs. Simulation results show that the proposed algorithm could not only guarantee the transmission rate of cellular users, but also improve the system and D2D pairs energy efficiency.

1 October 2021
Verified
Conference Paper
Closed access
SLSihan Liu
Sihan Liu
YWYue Wang
Yue Wang
Few-shot learning has been a long-standing problem in learning to learn. This problem typically involves training a model on an extremely small amount of data and testing the model on the out-of-distribution data. The focus of recent few-shot learning research has been on the development of good representation models that can quickly adapt to test tasks. To that end, we come up with a model that learns representation through online self-distillation. Our model combines supervised training with knowledge distillation via a continuously updated teacher. We also identify that data augmentation plays an important role in producing robust features. Our final model is trained with CutMix augmentation and online self-distillation. On the commonly used benchmark miniImageNet, our model achieves 67.07% and 83.03% under the 5-way 1-shot setting and the 5-way 5-shot setting, respectively. It outperforms counterparts of its kind by 2.25% and 0.89%.

24 February 2022 in Wireless Communications and Mobile Computing
Verified
Journal Article
Open access
PDF file
JSJihua Sheng
Jihua Sheng
SLSihan Liu
Sihan Liu
THTiancong Huang
Tiancong Huang
YWYucheng Wu
Yucheng Wu
Device-to-device (D2D) communication is one of the promising technologies for the next-generation cellular network, which uses direct communication between two neighboring devices to obtain a high transmission rate. This paper focuses on wireless resource allocation in a many-to-many D2D communication scenario to maximize the D2D transmission sum rate. Through the model analysis, the objective function is formulated while guaranteeing the minimum transmission rate of both the cellular users (CUs) and the D2D pairs. Based on the candidate sequence, an overlapping coalition game algorithm is proposed to enhance the D2D transmission sum rate. Furthermore, the preference sequence of CUs is adopted in the coalition initialization. According to the interference intensity, a power allocation scheme is designed further to improve the transmission sum rate of the D2D pairs. The simulation results have verified the validity of the proposed algorithm.

16 July 2022 in arXiv
Verified
Preprint
Open access
PDF file
DBDaniel Beaglehole
Daniel Beaglehole
MHMax Hopkins
Max Hopkins
DKDaniel Kane
Daniel Kane
SLSihan Liu
Sihan Liu
SLShachar Lovett
Shachar Lovett
Learning and equilibrium computation in games are fundamental problems across computer science and economics, with applications ranging from politics to machine learning. Much of the work in this area revolves around a simple algorithm termed \emph{randomized weighted majority} (RWM), also known as "Hedge" or "Multiplicative Weights Update," which is well known to achieve statistically optimal rates in adversarial settings (Littlestone and Warmuth '94, Freund and Schapire '99). Unfortunately, RWM comes with an inherent computational barrier: it requires maintaining and sampling from a distribution over all possible actions. In typical settings of interest the action space is exponentially large, seemingly rendering RWM useless in practice. In this work, we refute this notion for a broad variety of \emph{structured} games, showing it is possible to efficiently (approximately) sample the action space in RWM in \emph{polylogarithmic} time. This gives the first efficient no-regret algorithms for problems such as the \emph{(discrete) Colonel Blotto game}, \emph{matroid congestion}, \emph{matroid security}, and basic \emph{dueling games}. As an immediate corollary, we give a polylogarithmic time meta-algorithm to compute approximate Nash Equilibria for these games that is exponentially faster than prior methods in several important settings. Further, our algorithm is the first to efficiently compute equilibria for more involved variants of these games with general sums, more than two players, and, for Colonel Blotto, multiple resource types.