Author name cluster

Yu Lin

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

11 papers

2 author rows

AAAI Conference 2026 Conference Paper

S²Teacher: Step-by-step Teacher for Sparsely Annotated Oriented Object Detection

Yu Lin
Jianghang Lin
Kai Ye
You Shen
Shengchuan Zhang
Liujuan Cao

Although fully-supervised oriented object detection has made significant progress in remote sensing image understanding, it comes at the cost of labor-intensive annotation. Recent studies have explored weakly and semi-supervised learning to alleviate this burden. However, these methods overlook the difficulties posed by dense annotations in complex remote sensing scenes. In this paper, we introduce a novel setting called sparsely annotated oriented object detection (SAOOD), which only labels partial instances, and propose a solution to address its challenges. Specifically, we focus on two key issues in the setting: (1) sparse labeling leading to overfitting on limited foreground representations, and (2) unlabeled objects (false negatives) confusing feature learning. To this end, we propose the S2Teacher, a novel angle-consistency guided method that progressively mines pseudo-labels for unlabeled objects from easy to hard, enhancing foreground representations. Additionally, it reweights the loss of unlabeled objects to mitigate their impact during training. Extensive experiments demonstrate that S2Teacher not only significantly improves detector performance across different sparse annotation levels but also achieves near-fully-supervised performance on the DOTA dataset with only 10% annotation instances, effectively balancing accuracy and labeling cost.

PDF Details DOI

EAAI Journal 2025 Journal Article

A hybrid data-driven model for state of health estimation of Lithium-ion battery with capacity recovery

Yu Lin
Luo Zhou
Jianhai Yan
Shuguang He

The failure of lithium-ion batteries has attracted the attention of researchers. Monitoring the degradation process through sensor-based state-of-health (SOH) assessment enables early detection of anomalies and facilitates timely interventions to prevent catastrophic failures. During battery degradation, the phenomenon of capacity recovery makes it complicated to accurately evaluate the SOH. However, existing works usually neglect this phenomenon. In this paper, we propose a hybrid model focusing on capacity recovery. The model consists of four modules: a feature extraction module, a data decomposition module, a global trend assessment module, and a local fluctuation assessment module. Specifically, in the feature extraction module, a convolutional neural network extracts features from the input data that effectively characterize the capacity degradation process. The data decomposition module applies empirical mode decomposition (EMD) to separate local fluctuations caused by capacity recovery from the global degradation trend. These components are then input into their respective assessment modules. Additionally, the local fluctuation assessment module incorporates an attention mechanism that adaptively identifies the similarity between extracted features and health states, thereby enhancing the evaluation of local fluctuations. The proposed model is validated using the NASA (National Aeronautics and Space Administration) battery dataset, demonstrating its effectiveness in addressing the challenges of SOH prediction under capacity recovery conditions.

Details DOI

EAAI Journal 2025 Journal Article

Committee of Multi-scale Nonlinear Learning Frameworks for accurate stock price forecasting

Qian He
Yanhui Liang
Yu Lin
Dazhi Pan
Yuying Yue

Dependable stock price predictions are vital for optimizing economic policies and investment strategies in both national and corporate settings. However, the intrinsic volatility and intricacy of stock prices pose considerable challenges. Thus, this paper introduces a novel Committee of Multi-scale Nonlinear Learning Frameworks (CoML) that employs a three-stage model: decomposition, reconstruction, and prediction. First, a complete ensemble empirical mode decomposition with adaptive noise is adopted to decompose the original stock prices into multiple intrinsic mode functions. Secondly, a fine-to-coarse algorithm is applied to reconstruct the intrinsic mode functions, so as to effectively extract short-term fluctuations and long-term trends. Finally, an ensemble of nonlinear models including bidirectional long short-term memory (BiLSTM), support vector regression (SVR) and multi-layer perceptron (MLP) is used to learn and forecast features extracted to obtain high performance. Experimental results indicate that the model performs exceptionally well in both emerging and developed markets highlighting the innovative capabilities of CoML in highly complex and volatile financial markets. The proposed model is further validated using Model Confidence Set and the results indicate that the model is statistically significant.

Details DOI

ICLR Conference 2025 Conference Paper

Which Tasks Should Be Compressed Together? A Causal Discovery Approach for Efficient Multi-Task Representation Compression

Sha Guo
Jing Chen
Zixuan Hu
Zhuo Chen 0006
Wenhan Yang
Yu Lin
Xing Jiang
Lingyu Duan

Conventional image compression methods are inadequate for intelligent analysis, as they overemphasize pixel-level precision while neglecting semantic significance and the interaction among multiple tasks. This paper introduces a Taskonomy-Aware Multi-Task Compression framework comprising (1) inter-coherent task grouping, which organizes synergistic tasks into shared representations to improve multi-task accuracy and reduce encoding volume, and (2) a conditional entropy-based directed acyclic graph (DAG) that captures causal dependencies among grouped representations. By leveraging parent representations as contextual priors for child representations, the framework effectively utilizes cross-task information to improve entropy model accuracy. Experiments on diverse vision tasks, including Keypoint 2D, Depth Z-buffer, Semantic Segmentation, Surface Normal, Edge Texture, and Autoencoder, demonstrate significant bitrate-performance gains, validating the method’s capability to reduce system entropy uncertainty. These findings underscore the potential of leveraging representation disentanglement, synergy, and causal modeling to learn compact representations, which enable efficient multi-task compression in intelligent systems.

Details

ICRA Conference 2024 Conference Paper

SeqTrack3D: Exploring Sequence Information for Robust 3D Point Cloud Tracking

Yu Lin
Zhiheng Li 0003
Yubo Cui
Zheng Fang 0001

3D single object tracking (SOT) is an important and challenging task for the autonomous driving and mobile robotics. Most existing methods perform tracking between two consecutive frames while ignoring the motion patterns of the target over a series of frames, which would cause performance degradation in the scenes with sparse points. To break through this limitation, we introduce "Sequence-to-Sequence" tracking paradigm and a tracker named SeqTrack3D to capture target motion across continuous frames. Unlike previous methods that primarily adopted three strategies: matching two consecutive point clouds, predicting relative motion, or utilizing sequential point clouds to address feature degradation, our SeqTrack3D combines both historical point clouds and bounding box sequences. This novel method ensures robust tracking by leveraging location priors from historical boxes, even in scenes with sparse points. Extensive experiments conducted on large-scale datasets show that SeqTrack3D achieves new state-of-the-art performances, improving by 6. 00% on NuScenes and 14. 13% on Waymo dataset. The code will be made public at https://github.com/aron-lin/seqtrack3d.

Details

JBHI Journal 2022 Journal Article

An Algorithm to Mine Therapeutic Motifs for Cancer From Networks of Genetic Interactions

Herty Liany
Yu Lin
Anand Jeyasekharan
Vaibhav Rajan

Study of pairwise genetic interactions, such as mutually exclusive mutations, has led to understanding of underlying mechanisms in cancer. Investigation of various combinatorial motifs within networks of such interactions can lead to deeper insights into its mutational landscape and inform therapy development. One such motif called the Between-Pathway Model (BPM) represents redundant or compensatory pathways that can be therapeutically exploited. Finding such BPM motifs is challenging since most formulations require solving variants of the NP-complete maximum weight bipartite subgraph problem. In this paper we design an algorithm based on Integer Linear Programming (ILP) to solve this problem. In our experiments, our approach outperforms the best previous method to mine BPM motifs. Further, our ILP-based approach allows us to easily model additional application-specific constraints. We illustrate this advantage through a new application of BPM motifs that can potentially aid in finding combination therapies to combat cancer.

Details DOI

NeurIPS Conference 2022 Conference Paper

Graph Coloring via Neural Networks for Haplotype Assembly and Viral Quasispecies Reconstruction

Hansheng Xue
Vaibhav Rajan
Yu Lin

Understanding genetic variation, e. g. , through mutations, in organisms is crucial to unravel their effects on the environment and human health. A fundamental characterization can be obtained by solving the haplotype assembly problem, which yields the variation across multiple copies of chromosomes. Variations among fast evolving viruses that lead to different strains (called quasispecies) are also deciphered with similar approaches. In both these cases, high-throughput sequencing technologies that provide oversampled mixtures of large noisy fragments (reads) of genomes, are used to infer constituent components (haplotypes or quasispecies). The problem is harder for polyploid species where there are more than two copies of chromosomes. State-of-the-art neural approaches to solve this NP-hard problem do not adequately model relations among the reads that are important for deconvolving the input signal. We address this problem by developing a new method, called NeurHap, that combines graph representation learning with combinatorial optimization. Our experiments demonstrate the substantially better performance of NeurHap in real and synthetic datasets compared to competing approaches.

PDF Details

AAAI Conference 2022 Conference Paper

RepBin: Constraint-Based Graph Representation Learning for Metagenomic Binning

Hansheng Xue
Vijini Mallawaarachchi
Yujia Zhang
Vaibhav Rajan
Yu Lin

Mixed communities of organisms are found in many environments – from the human gut to marine ecosystems – and can have profound impact on human health and the environment. Metagenomics studies the genomic material of such communities through high-throughput sequencing that yields DNA subsequences for subsequent analysis. A fundamental problem in the standard workflow, called binning, is to discover clusters, of genomic subsequences, associated with the unknown constituent organisms. Inherent noise in the subsequences, various biological constraints that need to be imposed on them and the skewed cluster size distribution exacerbate the difficulty of this unsupervised learning problem. In this paper, we present a new formulation using a graph where the nodes are subsequences and edges represent homophily information. In addition, we model biological constraints providing heterophilous signal about nodes that cannot be clustered together. We solve the binning problem by developing new algorithms for (i) graph representation learning that preserves both homophily relations and heterophily constraints (ii) constraint-based graph clustering method that addresses the problems of skewed cluster size distribution. Extensive experiments, on real and synthetic datasets, demonstrate that our approach, called RepBin, outperforms a wide variety of competing methods. Our constraint-based graph representation learning and clustering methods, that may be useful in other domains as well, advance the state-of-the-art in both metagenomics binning and graph representation learning.

PDF Details

AAAI Conference 2021 Conference Paper

Single View Point Cloud Generation via Unified 3D Prototype

Yu Lin
Yigong Wang
Yi-Fan Li
Zhuoyi Wang
Yang Gao
Latifur Khan

As 3D point clouds become the representation of choice for multiple vision and graphics applications, such as autonomous driving, robotics, etc. , the generation of them by deep neural networks has attracted increasing attention in the research community. Despite the recent success of deep learning models in classification and segmentation, synthesizing point clouds remains challenging, especially from a single image. State-of-the-art (SOTA) approaches can generate a point cloud from a hidden vector, however, they treat 2D and 3D features equally and disregard the rich shape information within the 3D data. In this paper, we address this problem by integrating image features with 3D prototype features. Specifically, we propose to learn a set of 3D prototype features from a real point cloud dataset and dynamically adjust them through the training. These prototypes are then integrated with incoming image features to guide the point cloud generation process. Experimental results show that our proposed method outperforms SOTA methods on single image based 3D reconstruction tasks.

PDF Details

AAAI Conference 2020 Short Paper

Few Sample Learning without Data Storage for Lifelong Stream Mining (Student Abstract)

Zhuoyi Wang
Yigong Wang
Yu Lin
Bo Dong
Hemeng Tao
Latifur Khan

Continuously mining complexity data stream has recently been attracting an increasing amount of attention, due to the rapid growth of real-world vision/signal applications such as self-driving cars and online social media messages. In this paper, we aim to address two signiﬁcant problems in the lifelong/incremental stream mining scenario: ﬁrst, how to make the learning algorithms generalize to the unseen classes only from a few labeled samples; second, is it possible to avoid storing instances from previously seen classes to solve the catastrophic forgetting problem? We introduce a novelty stream mining framework to classify the inﬁnite stream of data with different categories that occurred during different times. We apply a few-sample learning strategy to make the model recognize the novel class with limited samples; at the same time, we implement an incremental generative model to maintain old knowledge when learning new coming categories, and also avoid the violation of data privacy and memory restrictions simultaneously. We evaluate our approach in the continual class-incremental setup on the classiﬁcation tasks and ensure the sufﬁcient model capacity to accommodate for learning the new incoming categories.

PDF Details

TCS Journal 2018 Journal Article

Can a breakpoint graph be decomposed into none other than 2-cycles?

Lianrong Pu
Yu Lin
Daming Zhu
Haitao Jiang

Breakpoint graph has been widely used as a key data structure in algorithm design for genome rearrangements. The problem of breakpoint graph cycle decomposition, which asks for a largest collection of edge-disjoint cycles, is crucial in computing rearrangement distances between genomes. This problem is NP-hard, and can be approximated to 1. 4193 + ϵ. It is still open for deciding whether a breakpoint graph can admit a cycle decomposition with none other than 2-cycles. In this paper, we present a linear time algorithm to detect whether a breakpoint graph can be decomposed into none other than 2-cycles.

Details DOI