1 code implementation • 21 Apr 2024 • Yihao Zhang, Zeming Wei, Jun Sun, Meng Sun
Recent research has introduced Representation Engineering (RepE) as a promising approach for understanding complex inner workings of large-scale models like Large Language Models (LLMs).
no code implementations • 2 Apr 2024 • Qizhi Qiu, Xiaomin Liu, Yihao Zhang, Lilin Yi, Weisheng Hu, Qunbi Zhuge
We propose a heuristic-based optimization scheme for reliable optical amplifier reconfiguration process in ADON.
1 code implementation • 23 Feb 2024 • Yihao Zhang, Hangzhou He, Jingyu Zhu, Huanran Chen, Yifei Wang, Zeming Wei
Instead of perturbing the samples, Sharpness-Aware Minimization (SAM) perturbs the model weights during training to find a more flat loss landscape and improve generalization.
no code implementations • 13 Jul 2023 • Yichen Liu, Xiaomin Liu, Yihao Zhang, Meng Cai, Mengfan Fu, Xueying Zhong, Lilin Yi, Weisheng Hu, Qunbi Zhuge
To enable intelligent and self-driving optical networks, high-accuracy physical layer models are required.
1 code implementation • 24 Jun 2023 • Zeming Wei, Xiyue Zhang, Yihao Zhang, Meng Sun
In this paper, we propose a novel framework of Weighted Finite Automata (WFA) extraction and explanation to tackle the limitations for natural language tasks.
1 code implementation • 9 May 2023 • Zeming Wei, Jingyu Zhu, Yihao Zhang
In this paper, we explore SAM in the context of adversarial robustness.
1 code implementation • 20 Apr 2023 • Yihao Zhang, Zeming Wei, Xiyue Zhang, Meng Sun
To evaluate the effectiveness of our implementation and improvements, we conduct extensive experiments on a set of benchmark datasets.
no code implementations • 24 Jun 2022 • Yihao Zhang, Xiaomin Liu, Yichen Liu, Lilin Yi, Weisheng Hu, Qunbi Zhuge
Based on the physical features of Raman amplification, we propose a three-step modelling scheme based on neural networks (NN) and linear regression.
no code implementations • 13 Jun 2022 • Xiaomin Liu, Yuli Chen, Yihao Zhang, Yichen Liu, Lilin Yi, Weisheng Hu, Qunbi Zhuge
We propose a physics-informed EDFA gain model based on the active learning method.
no code implementations • 1 Apr 2021 • Yihao Zhang, John J. Leonard
Recent achievements in depth prediction from a single RGB image have powered the new research area of combining convolutional neural networks (CNNs) with classical simultaneous localization and mapping (SLAM) algorithms.
no code implementations • 19 Mar 2021 • Yihao Zhang, John J. Leonard
For a robot deployed in the world, it is desirable to have the ability of autonomous learning to improve its initial pre-set knowledge.
no code implementations • 30 Nov 2020 • Yihao Zhang, Zhaojie Chai, George Lykotrafitis
Overall, we show that our model can efficiently simulate emergency evacuation in complex environments with multiple room exits and obstacles where it is difficult to obtain an intuitive rule for fast evacuation.
2 code implementations • 22 Oct 2020 • Hascoet Tristan, Yihao Zhang, Persch Andreas, Ryoichi Takashima, Tetsuya Takiguchi, Yasuo Ariki
Maintaining aging infrastructure is a challenge currently faced by local and national administrators all around the world.
no code implementations • 7 Jul 2020 • Yihao Zhang, Zhaojie Chai, Yubing Sun, George Lykotrafitis
Because of the different migration mechanisms of leader and follower neural crest cells, we train two types of agents (leaders and followers) to learn the collective cell migration behavior.
no code implementations • 10 Feb 2018 • Chuanyun Xu, Yang Zhang, Xin Feng, YongXing Ge, Yihao Zhang, Jianwu Long
We focus on one-shot classification by deep learning approach based on a small quantity of training samples.