no code implementations • 22 Aug 2024 • Mahdi Kazemi, Aftab Hussain, Md Rafiqul Islam Rabin, Mohammad Amin Alipour, Sen Lin
This work investigates the application of Machine Unlearning (MU) for mitigating the impact of trojans embedded in conventional large language models of natural language (Text-LLMs) and large language models of code (Code-LLMs) We propose a novel unlearning approach, LYA, that leverages both gradient ascent and elastic weight consolidation, a Fisher Information Matrix (FIM) based regularization technique, to unlearn trojans from poisoned models.
no code implementations • 24 Jun 2024 • Hongbo Li, Sen Lin, Lingjie Duan, Yingbin Liang, Ness B. Shroff
Continual learning (CL) has garnered significant attention because of its ability to adapt to new tasks that arrive over time.
1 code implementation • 24 May 2024 • Sheng Yue, Jiani Liu, Xingyuan Hua, Ju Ren, Sen Lin, Junshan Zhang, Yaoxue Zhang
Offline Imitation Learning (IL) with imperfect demonstrations has garnered increasing attention owing to the scarcity of expert data in many real-world domains.
1 code implementation • 24 May 2024 • Sheng Yue, Xingyuan Hua, Ju Ren, Sen Lin, Junshan Zhang, Yaoxue Zhang
In this paper, we study offline-to-online Imitation Learning (IL) that pretrains an imitation policy from static demonstration data, followed by fast finetuning with minimal environmental interaction.
no code implementations • 23 Jan 2024 • Yun Peng, Sen Lin, Qian Chen, Lyu Xu, Xiaojun Ren, Yafei Li, Jianliang Xu
Graph analysis is fundamental in real-world applications.
no code implementations • 9 Jan 2024 • Shengli Zhang, Zhiyong Tao, Sen Lin
However, the haze distribution in the real world is complex, and downsampling can lead to color distortion or loss of detail in the output results as the resolution of a feature map or image resolution decreases.
no code implementations • 9 Jan 2024 • Shengli Zhang, Zhiyong Tao, Sen Lin
Compared to existing state-of-the-art methods, DedustNet achieves superior performance and more reliable results in agricultural image dedusting, providing strong support for the application of agricultural machinery in dusty environments.
no code implementations • 18 Aug 2023 • Yung-Fu Chen, Sen Lin, Anish Arora
We propose a learning algorithm for local routing policies that needs only a few data samples obtained from a single graph while generalizing to all random graphs in a standard model of wireless networks.
no code implementations • 1 Aug 2023 • Daouda Sow, Sen Lin, Zhangyang Wang, Yingbin Liang
Experiments on standard classification datasets demonstrate that our proposed approach outperforms related state-of-the-art baseline methods in terms of average robust performance, and at the same time improves the robustness against attacks on the weakest data points.
no code implementations • 21 Jul 2023 • Viraj Mehta, Ojash Neopane, Vikramjeet Das, Sen Lin, Jeff Schneider, Willie Neiswanger
Preference-based feedback is important for many applications where direct evaluation of a reward function is not feasible.
no code implementations • NeurIPS 2021 • Hang Wang, Sen Lin, Junshan Zhang
It is known that the estimation bias hinges heavily on the ensemble size (i. e., the number of Q-function approximators used in the target), and that determining the `right' ensemble size is highly nontrivial, because of the time-varying nature of the function approximation errors during the learning process.
no code implementations • 20 Jun 2023 • Hang Wang, Sen Lin, Junshan Zhang
To this end, the primary objective of this work is to build a fundamental understanding on ``\textit{whether and when online learning can be significantly accelerated by a warm-start policy from offline RL?}''.
no code implementations • 8 Jun 2023 • Peizhong Ju, Sen Lin, Mark S. Squillante, Yingbin Liang, Ness B. Shroff
For example, when the total number of features in the source task's learning model is fixed, we show that it is more advantageous to allocate a greater number of redundant features to the task-specific part rather than the common part.
no code implementations • 13 Mar 2023 • Li Yang, Sen Lin, Fan Zhang, Junshan Zhang, Deliang Fan
Inspired by the success of Self-supervised learning (SSL) in learning visual representations from unlabeled data, a few recent works have studied SSL in the context of continual learning (CL), where multiple tasks are learned sequentially, giving rise to a new paradigm, namely self-supervised continual learning (SSCL).
no code implementations • 12 Feb 2023 • Sen Lin, Peizhong Ju, Yingbin Liang, Ness Shroff
In particular, there is a lack of understanding on what factors are important and how they affect "catastrophic forgetting" and generalization performance.
no code implementations • 9 Feb 2023 • Sheng Yue, Guanbo Wang, Wei Shao, Zhaofeng Zhang, Sen Lin, Ju Ren, Junshan Zhang
This work aims to tackle a major challenge in offline Inverse Reinforcement Learning (IRL), namely the reward extrapolation error, where the learned reward function may fail to explain the task correctly and misguide the agent in unseen environments due to the intrinsic covariate shift.
no code implementations • 2 Feb 2023 • Daouda Sow, Sen Lin, Yingbin Liang, Junshan Zhang
More specifically, we first propose two simple but effective detection mechanisms of task switches and distribution shift based on empirical observations, which serve as a key building block for more elegant online model updates in our algorithm: the task switch detection mechanism allows reusing of the best model available for the current task at hand, and the distribution shift detection mechanism differentiates the meta model update in order to preserve the knowledge for in-distribution tasks and quickly learn the new knowledge for out-of-distribution tasks.
no code implementations • 1 Nov 2022 • Sen Lin, Li Yang, Deliang Fan, Junshan Zhang
By learning a sequence of tasks continually, an agent in continual learning (CL) can improve the learning performance of both a new task and `old' tasks by leveraging the forward knowledge transfer and the backward knowledge transfer, respectively.
1 code implementation • ICLR 2022 • Sen Lin, Li Yang, Deliang Fan, Junshan Zhang
To tackle this challenge, we propose Trust Region Gradient Projection (TRGP) for continual learning to facilitate the forward knowledge transfer based on an efficient characterization of task correlation.
no code implementations • ICLR 2022 • Sen Lin, Jialin Wan, Tengyu Xu, Yingbin Liang, Junshan Zhang
In particular, we devise a new meta-Regularized model-based Actor-Critic (RAC) method for within-task policy optimization, as a key building block of MerPO, using conservative policy evaluation and regularized policy improvement; and the intrinsic tradeoff therein is achieved via striking the right balance between two regularizers, one based on the behavior policy and the other on the meta-policy.
1 code implementation • 1 Feb 2022 • Liang Liao, Sen Lin, Lun Li, Xiuwei Zhang, Song Zhao, Yan Wang, Xinqiang Wang, Qi Gao, Jingyu Wang
Higher order singular value decomposition (HOSVD) extends the SVD and can approximate higher order data using sums of a few rank-one components.
no code implementations • 3 Oct 2021 • Li Yang, Sen Lin, Junshan Zhang, Deliang Fan
To address this issue, continual learning has been developed to learn new tasks sequentially and perform knowledge transfer from the old tasks to the new ones without forgetting.
no code implementations • 22 Jan 2021 • Mehmet Dedeoglu, Sen Lin, Zhaofeng Zhang, Junshan Zhang
Learning generative models is challenging for a network edge node with limited data and computing power.
1 code implementation • 17 Jan 2021 • Liang Liao, Xuechun Zhang, Xinqiang Wang, Sen Lin, Xin Liu
We also show in our experiments that the performance of TPCA increases when the order of compounded pixels increases.
no code implementations • 22 Dec 2020 • Hang Wang, Sen Lin, Hamid Jafarkhani, Junshan Zhang
Specifically, we assume that agents maintain local estimates of the global state based on their local information and communications with neighbors.
no code implementations • 16 Dec 2020 • Sheng Yue, Ju Ren, Jiang Xin, Sen Lin, Junshan Zhang
To overcome these challenges, we explore continual edge learning capable of leveraging the knowledge transfer from previous tasks.
no code implementations • 15 Dec 2020 • Sen Lin, Mehmet Dedeoglu, Junshan Zhang
By characterizing the upper bound of the agent-task-averaged regret, we show that the performance of multi-agent online meta-learning depends heavily on how much an agent can benefit from the distributed network-level OCO for meta-model updates via limited communication, which however is not well understood.
no code implementations • 25 Nov 2020 • Sen Lin, Li Yang, Zhezhi He, Deliang Fan, Junshan Zhang
In this work, we advocate a holistic approach to jointly train the backbone network and the channel gating which enables dynamical selection of a subset of filters for more efficient local computation given the data input.
no code implementations • 27 Oct 2020 • Sen Lin, Hang Wang, Junshan Zhang
System identification is a fundamental problem in reinforcement learning, control theory and signal processing, and the non-asymptotic analysis of the corresponding sample complexity is challenging and elusive, even for linear time-varying (LTV) systems.
no code implementations • 11 Apr 2020 • Sen Lin, Kaichen Chi
Firstly, the color equalization of the degraded image is realized by the automatic color enhancement algorithm; Secondly, the relative total variation is introduced to decompose the image into the structure layer and texture layer; Then, the best background light point is selected based on brightness, gradient discrimination, and hue judgment, the transmittance of the backscatter component is obtained by the red dark channel prior, which is substituted into the imaging model to remove the fogging phenomenon in the structure layer.
no code implementations • 9 Jan 2020 • Sen Lin, Guang Yang, Junshan Zhang
Further, we investigate the convergence of the proposed federated meta-learning algorithm under mild conditions on node similarity and the adaptation performance at the target edge.
no code implementations • 24 Jul 2019 • Shaodi You, Erqi Huang, Shuaizhe Liang, Yongrong Zheng, Yunxiang Li, Fan Wang, Sen Lin, Qiu Shen, Xun Cao, Diming Zhang, Yuanjiang Li, Yu Li, Ying Fu, Boxin Shi, Feng Lu, Yinqiang Zheng, Robby T. Tan
This document introduces the background and the usage of the Hyperspectral City Dataset and the benchmark.