Yi Chen

AAAI Conference 2026 Conference Paper

CD-DPE: Dual-Prompt Expert Network Based on Convolutional Dictionary Feature Decoupling for Multi-Contrast MRI Super-Resolution

• Xianming Gu
• Lihui Wang
• Ying Cao
• Zeyu Deng
• Yingfeng Ou
• Guodong Hu
• Yi Chen

Multi-contrast magnetic resonance imaging (MRI) super-resolution intends to reconstruct high-resolution (HR) images from low-resolution (LR) scans by leveraging structural information present in HR reference images acquired with different contrasts. This technique enhances anatomical detail and soft tissue differentiation, which is vital for early diagnosis and clinical decision-making. However, inherent contrasts disparities between modalities pose fundamental challenges in effectively utilizing reference image textures to guide target image reconstruction, often resulting in suboptimal feature integration. To address this issue, we propose a dual-prompt expert network based on a convolutional dictionary feature decoupling (CD-DPE) strategy for multi-contrast MRI super-resolution. Specifically, we introduce an iterative convolutional dictionary feature decoupling module (CD-FDM) to separate features into cross-contrast and intra-contrast components, thereby reducing redundancy and interference. To fully integrate these features, a novel dual-prompt feature fusion expert module (DP-FFEM) is proposed. This module uses a frequency prompt to guide the selection of relevant reference features for incorporation into the target image, while an adaptive routing prompt determines the optimal method for fusing reference and target features to enhance reconstruction quality. Extensive experiments on public multi-contrast MRI datasets demonstrate that CD-DPE outperforms state-of-the-art methods in reconstructing fine details. Additionally, experiments on unseen datasets demonstrated that CD-DPE exhibits strong generalization capabilities.

AAAI Conference 2026 Conference Paper

Not Just What’s There: Enabling CLIP to Comprehend Negated Visual Descriptions Without Fine-Tuning

• Junhao Xiao
• Zhiyu Wu
• Hao Lin
• Yi Chen
• Yahui Liu
• Xiaoran Zhao
• Zixu Wang
• Zejiang He

Vision-Language Models (VLMs) like CLIP struggle to understand negation, often embedding affirmatives and negatives similarly (e.g., matching "no dog" with dog images). Existing methods refine negation understanding via fine-tuning CLIP’s text encoder, risking overfitting. In this work, we propose CLIPGlasses, a plug-and-play framework that enhances CLIP’s ability to comprehend negated visual descriptions. CLIPGlasses adapts a dual-stage design: a Lens module disentangles negated semantics from text embeddings, and a Frame module predicts context-aware repulsion strength, which is integrated into the modified similarity computation to penalize alignment with negated semantics, thereby reducing false positive matches. Experiments show that CLIP equipped with CLIPGlasses achieves competitive in-domain performance and outperforms state-of-the-art methods in cross-domain generalization. Its superiority is especially evident under low-resource conditions, indicating stronger robustness across domains.

AAAI Conference 2026 Conference Paper

ROVER: Robust Generative Continual Identity Unlearning Against Relearning Attacks

• Tairan Huang
• Qiang Chen
• Beibei Hu
• Yunlong Zhao
• Hongyan Xu
• Zhiyuan Chen
• Yi Chen
• Xiu Su

Recent generative unlearning models synthesize high quality samples while protecting private information by unlearning the identity. However, existing generative identity unlearning methods face two challenges in multi-identity unlearning: 1) identity conflicts, which cause conflicts of model parameters in the continuous erasure of multiple identities; 2) fragile unlearning, where the model's unlearning ability deteriorates or fails under malicious attacks. In this paper, we introduce a critical yet under-explored task called robust multi-identity unlearning, with the goals of resolving identity conflicts to achieve interference-free unlearning and protecting against malicious attacks to achieve robust unlearning. To satisfy these goals, we propose a novel framework, RObust generatiVE continual identity unlearning against Relearning attacks (ROVER). By filtering unlearning requests with latent similarity, our method effectively isolates benign unlearning from malicious attacks to preserve identity removal integrity. Meanwhile, residual orthogonal resonator resolves identity conflicts in the continuous erasure of multiple identities, preserving stability in benign continual unlearning. Moreover, we introduce the phantom guard network to block malicious attacks by absorbing adversarial gradients, ensuring irreversible identity unlearning. The extensive experiments demonstrate that our proposed method achieves state-of-the-art performance on the task of robust multi-identity unlearning against relearning attacks.

AAAI Conference 2026 Conference Paper

VAGU & GtS: LLM-Based Benchmark and Framework for Joint Video Anomaly Grounding and Understanding

• Shibo Gao
• Peipei Yang
• Yangyang Liu
• Yi Chen
• Han Zhu
• Xu-Yao Zhang
• Linlin Huang

For video anomaly detection, it's both important to detect when the event happens and what the event is. The tasks of temporal grounding and semantic understanding can benefit from joint learning, but no existing work support it. To address this problem, we introduce VAGU (Video Anomaly Grounding and Understanding), the first benchmark designed to jointly evaluate semantic understanding and precise temporal grounding of anomalies, with comprehensive annotations and objective multiple-choice Video QA. Besides, we propose Glance then Scrutinize (GtS), the first training-free framework that achieves the best balance performance in both accuracy and efficiency. GtS uniquely balances high temporal precision and semantic interpretability while meeting practical speed requirements, outperforming previous methods in real-world scenarios. Furthermore, we introduce the JeAUG metric for holistic evaluation of both speed and accuracy. Extensive experiments demonstrate the superior effectiveness and practicality of our benchmark, framework, and metric.

AAAI Conference 2025 Conference Paper

Aerodynamic Coefficients Prediction via Cross-Attention Fusion and Physical-Informed Training

• Yueqing Wang
• Peng Zhang
• Yushuang Liu
• Jianing Zhao
• Jie Lin
• Yi Chen

Aerodynamic coefficient prediction is pivotal in aircraft and vehicles' design, performance evaluation, and motion control. Integrating artificial neural networks into aerodynamic coefficient prediction offers a promising alternative to traditional numerical methods burdened by extensive computations and high costs. Nevertheless, this data-driven approach faces several critical challenges, which limit its further performance enhancement: i) The current research lacks a profound understanding of the complex interplay between the shape of an object and its aerodynamic characteristics. ii) The scarcity of high-quality aerodynamic data poses a significant barrier. The models trained on limited datasets lack generalization ability, struggling to accurately predict and adapt to diverse aerodynamic performance under new shapes or conditions. To overcome these challenges, we introduce an innovative framework that employs cross-attention to capture the intimate interplay between shape and flow conditions and allows for the direct utilization of pre-trained models on general shape datasets to mitigate the scarcity of aerodynamic data. Furthermore, to bolster the inference capabilities of this data-driven approach, we integrate physical information constraints into the model, leveraging them as guiding principles to enhance the model's predictive power under unknown conditions. Experimental validation demonstrates that our proposed method performs excellently in multiple aerodynamic prediction tasks. This achievement brings a new technological breakthrough to the field of aerodynamic prediction and provides robust support for the design optimization of complex systems such as aircraft and vehicles.

AAAI Conference 2025 Conference Paper

AoP-SAM: Automation of Prompts for Efficient Segmentation

• Yi Chen
• Muyoung Son
• Chuanbo Hua
• Joo-Young Kim

The Segment Anything Model (SAM) is a powerful foundation model for image segmentation, showing robust zero-shot generalization through prompt engineering. However, relying on manual prompts is impractical for real-world applications, particularly in scenarios where rapid prompt provision and resource efficiency are crucial. In this paper, we propose the Automation of Prompts for SAM (AoP-SAM), a novel approach that learns to generate essential prompts in optimal locations automatically. AoP-SAM enhances SAM’s efficiency and usability by eliminating manual input, making it better suited for real-world tasks. Our approach employs a lightweight yet efficient Prompt Predictor model that detects key entities across images and identifies the optimal regions for placing prompt candidates. This method leverages SAM’s image embeddings, preserving its zero-shot generalization capabilities without requiring fine-tuning. Additionally, we introduce a test-time instance-level Adaptive Sampling and Filtering mechanism that generates prompts in a coarse-to-fine manner. This notably enhances both prompt and mask generation efficiency by reducing computational overhead and minimizing redundant mask refinements. Evaluations of three datasets demonstrate that AoP-SAM substantially improves both prompt generation efficiency and mask generation accuracy, making SAM more effective for automated segmentation tasks.

ICLR Conference 2025 Conference Paper

Decoupling Layout from Glyph in Online Chinese Handwriting Generation

• Minsi Ren
• Yan-Ming Zhang
• Yi Chen

Text plays a crucial role in the transmission of human civilization, and teaching machines to generate online handwritten text in various styles presents an interesting and significant challenge. However, most prior work has concentrated on generating individual Chinese fonts, leaving complete text line generation largely unexplored. In this paper, we identify that text lines can naturally be divided into two components: layout and glyphs. Based on this division, we designed a text line layout generator coupled with a diffusion-based stylized font synthesizer to address this challenge hierarchically. More concretely, the layout generator performs in-context-like learning based on the text content and the provided style references to generate positions for each glyph autoregressively. Meanwhile, the font synthesizer which consists of a character embedding dictionary, a multi-scale calligraphy style encoder and a 1D U-Net based diffusion denoiser will generate each font on its position while imitating the calligraphy style extracted from the given style references. Qualitative and quantitative experiments on the CASIA-OLHWDB demonstrate that our method is capable of generating structurally correct and indistinguishable imitation samples.

EAAI Journal 2025 Journal Article

Inter-layer explainable variational autoencoder model for multivariate time series anomaly detection

• Xiaoxia Zhang
• Guangyao Wang
• Yi Chen
• Wenzhi Yang
• Guoyin Wang

The rapid expansion of big data and artificial intelligence has significantly increased the dimensionality and quantity of monitoring data across various industries. Multivariate time series data, one of the most frequently used data types in various industries, often contains anomalies caused by human error or electromagnetic interference. Detecting these anomalies is essential, the volume of data makes manual labeling impractical, and supervised methods are limited by their inability to address unknown anomalies. Additionally, existing deep learning detection methods often function as black-box models, reducing trust in the results and obscuring the testing process. To address these challenges, we propose Shapley Attention Variational Autoencoder (SHAPAttenVAE), an inter-layer explainable variational autoencoder (VAE) model that combines shapley values with a multi-head attention mechanism. SHAPAttenVAE leverages shapley values to quantify the contribution of nodes from previous layers to their successors, helping to identify root causes of anomalies across dimensions. The model is optimized using a two-phase training method inspired by adversarial network, where random Gaussian noise is introduced to reduce noise interference within the data. Experimental results on several large datasets demonstrate that SHAPAttenVAE outperforms existing models in both accuracy and robustness. These findings establish SHAPAttenVAE as an effective and interpretable framework for anomaly detection in multivariate time series, making it a promising tool for widespread industry application.

AAAI Conference 2025 Conference Paper

Learning Verified Safe Neural Network Controllers for Multi-Agent Path Finding

• Mingyue Zhang
• Nianyu Li
• Yi Chen
• Jialong Li
• Xiao-Yi Zhang
• Hengjun Zhao
• Jiamou Liu
• Wu Chen

Multi-agent path finding (MAPF) is a safety-critical scenario where the goal is to secure collision-free trajectories from initial to desired locations. However, due to system complexity and uncertainty, integrating learning-based controllers with MAPF is challenging and cannot theoretically guarantee the safety of the learned controllers. In response, our study proposes a verified safe multi-agent neural control (VSMANC) approach for MAPF, focusing on the unified training of Decentralized Control Barrier Functions (DCBF) and controllers to enhence safety. VSMANC enables all agents to concurrently learn controllers and DCBFs using a unified loss function designed to maximize safety, adhere to standard control policies, and incorporate path-finding-related heuristics. We also propose a formal verification-guided retraining process to both verify the properties of the learned DCBFs and generate counterexamples for retraining, thereby providing a verified safety guarantee. We validate our approach through shape formation experiments and UAV simulations, demonstrating significant improvements in safety and effectiveness in complex multi-agent environments.

UAI Conference 2025 Conference Paper

Metric Learning in an RKHS

• Gokcan Tatli
• Yi Chen
• Blake Mason
• Robert D. Nowak
• Ramya Korlakai Vinayak

This paper investigates metric learning in a Reproducing Kernel Hilbert Space (RKHS) based on a set of random triplet comparisons in the form of *"Do you think item h is more similar to item i or item j? "* indicating similarity and differences between various items. The goal is to learn a metric in the RKHS that reflects the comparisons. Nonlinear metric learning using kernel methods and neural networks has shown great empirical promise. While previous works have addressed certain aspects of this problem, there is little or no theoretical understanding of such methods. The exception is the special (linear) case in which the RKHS is the standard $d$-dimensional Euclidean space; there is a comprehensive theory for metric learning in the $d$-dimensional Euclidean space. This paper develops a general RKHS framework for metric learning and provides novel generalization guarantees and sample complexity bounds. We validate our findings through a set of simulations and experiments on real datasets. Our code is publicly available at https: //github. com/RamyaLab/metric-learning-RKHS.

ICLR Conference 2025 Conference Paper

PAL: Sample-Efficient Personalized Reward Modeling for Pluralistic Alignment

• Daiwei Chen
• Yi Chen
• Aniket Rege
• Zhi Wang
• Ramya Korlakai Vinayak

Foundation models trained on internet-scale data benefit from extensive alignment to human preferences before deployment. However, existing methods typically assume a homogeneous preference shared by all individuals, overlooking the diversity inherent in human values. In this work, we propose a general reward modeling framework for pluralistic alignment (PAL), which incorporates diverse preferences from the ground up. PAL has a modular design that leverages commonalities across users while catering to individual personalization, enabling efficient few-shot localization of preferences for new users. Extensive empirical evaluation demonstrates that PAL matches or outperforms state-of-the-art methods on both text-to-text and text-to-image tasks: on Reddit TL;DR Summary, PAL is 1.7% more accurate for seen users and 36% more accurate for unseen users compared to the previous best method, with 100× less parameters. On Pick-a-Pic v2, PAL is 2.5% more accurate than the best method with 156× fewer learned parameters. Finally, we provide theoretical analysis for generalization of rewards learned via PAL framework showcasing the reduction in number of samples needed per user.

AAAI Conference 2025 Conference Paper

Recoverable Compression: A Multimodal Vision Token Recovery Mechanism Guided by Text Information

• Yi Chen
• Jian Xu
• Xu-Yao Zhang
• Wen-Zhuo Liu
• Yang-Yang Liu
• Cheng-Lin Liu

With the advancement of large-scale language modeling techniques, large multimodal models combining visual encoders with large language models have demonstrated exceptional performance in various visual tasks. Most of the current large multimodal models achieve this by mapping visual features obtained from the visual encoder into a large language model and using them as inputs alongside text for downstream tasks. Therefore, the number of visual tokens directly affects the training and inference speed of the model. There has been significant work on token pruning for visual transformers, but for large multimodal models, only relying on visual information for token pruning or compression may lead to significant loss of important information. On the other hand, the textual input in the form of a question may contain valuable information that can aid in answering the question, providing additional knowledge to the model. To address the potential oversimplification and excessive pruning that can occur with most purely visual token pruning methods, we propose a text information-guided dynamic visual token recovery mechanism that does not require training. This mechanism leverages the similarity between the question text and visual tokens to recover visually meaningful tokens with important text information while merging other less important tokens, to achieve efficient computation for large multimodal models. Experimental results demonstrate that our proposed method achieves comparable performance to the original approach while compressing the visual tokens to an average of 10\% of the original quantity.

ICML Conference 2025 Conference Paper

Rethinking Confidence Scores and Thresholds in Pseudolabeling-based SSL

• Harit Vishwakarma
• Yi Chen
• Satya Sai Srinath Namburi GNVV
• Sui Jiet Tay
• Ramya Korlakai Vinayak
• Frederic Sala

Modern semi-supervised learning (SSL) methods rely on pseudolabeling and consistency regularization. Pseudolabeling is typically performed by comparing the model’s confidence scores and a predefined threshold. While several heuristics have been proposed to improve threshold selection, the underlying issues of overconfidence and miscalibration in confidence scores remain largely unaddressed, leading to inaccurate pseudolabels, degraded test accuracy, and prolonged training. We take a first-principles approach to learn confidence scores and thresholds with an explicit knob for error. This flexible framework addresses the fundamental question of optimal scores and threshold selection in pseudolabeling. Moreover, it gives practitioners a principled way to control the quality and quantity of pseudolabels. Such control is vital in SSL, where balancing pseudolabel quality and quantity directly affects model performance and training efficiency. Our experiments show that, by integrating this framework with modern SSL methods, we achieve significant improvements in accuracy and training efficiency. In addition, we provide novel insights on the trade-offs between the choices of the error parameter and the end model’s performance.

AIIM Journal 2025 Journal Article

Rough hypervolume-driven feature selection with groupwise intelligent sampling for detecting clinical characterization of lupus nephritis

• Xinsen Zhou
• Yi Chen
• Ali Asghar Heidari
• Huiling Chen
• Xiaowei Chen

Systemic lupus erythematosus (SLE) is an autoimmune inflammatory disease. Lupus nephritis (LN) is a major risk factor for morbidity and mortality in SLE. Proliferative and pure membranous LN have different prognoses and may require different treatments. This study proposes a binary rough hypervolume-driven spherical evolution algorithm with groupwise intelligent sampling (bRGSE). The efficient dimensionality reduction capability of the bRGSE is verified across twelve datasets. These datasets are from the public datasets, with feature dimensions ranging from seven hundred to fifty thousand. The experimental results indicate that bRGSE performs better than seven high-performing alternatives. Then, the bRGSE was combined with adaptive boosting (AdaBoost) to form a new model (bRGSE_AdaBoost), which analyzed clinical records collected from 110 patients with LN. Experimental results show that the proposed bRGSE_AdaBoost can identify the most critical indicators, including urine latent blood, white blood cells, endogenous creatinine clearing rate, and age. These indicators may help differentiate between proliferative LN and membranous LN. The proposed bRGSE algorithm is an efficient dimensionality reduction method. The developed bRGSE_AdaBoost model, a computer-aided model, achieved an accuracy of 96. 687 % and is expected to provide early warning for the treatment and diagnosis of LN.

IROS Conference 2025 Conference Paper

SN-LiDAR: Semantic Neural Fields for Novel Space-time View LiDAR Synthesis

• Yi Chen
• Tianchen Deng
• Wentao Zhao
• Xiaoning Wang
• Wenqian Xi
• Weidong Chen
• Jingchuan Wang

Recent research has begun exploring novel view synthesis (NVS) for LiDAR point clouds, aiming to generate realistic LiDAR scans from unseen viewpoints. However, most existing approaches do not reconstruct semantic labels, which are crucial for many downstream applications such as autonomous driving and robotic perception. Unlike images, which benefit from powerful segmentation models, LiDAR point clouds lack such large-scale pre-trained models, making semantic annotation time-consuming and labor-intensive. To address this challenge, we propose SN-LiDAR, a method that jointly performs accurate semantic segmentation, high-quality geometric reconstruction, and realistic LiDAR synthesis. Specifically, we employ a coarse-to-fine planar-grid feature representation to extract global features from multi-frame point clouds and leverage a CNN-based encoder to extract local semantic features from the current frame point cloud. Extensive experiments on SemanticKITTI and KITTI-360 demonstrate the superiority of SN-LiDAR in both semantic and geometric reconstruction, effectively handling dynamic objects and large-scale scenes. Codes will be available on https://github.com/dtc111111/SN-Lidar.

IROS Conference 2025 Conference Paper

TOPP-DWR: Time-Optimal Path Parameterization of Differential-Driven Wheeled Robots Considering Piecewise-Constant Angular Velocity Constraints

• Yong Li
• Yujun Huang
• Yi Chen
• Hui Cheng

Differential-driven wheeled robots (DWR) represent the quintessential type of mobile robots and find extensive applications across the robotic field. Most high-performance control approaches for DWR explicitly utilize the linear and angular velocities of the trajectory as control references. However, existing research on time-optimal path parameterization (TOPP) for mobile robots usually neglects the angular velocity and joint velocity constraints, which can result in degraded control performance in practical applications. In this article, a systematic and practical TOPP algorithm named TOPP-DWR is proposed for DWR and other mobile robots. First, the non-uniform B-spline is adopted to represent the initial trajectory in the task space. Second, the piecewise-constant angular velocity, as well as joint velocity, linear velocity, and linear acceleration constraints, are incorporated into the TOPP problem. During the construction of the optimization problem, the aforementioned constraints are uniformly represented as linear velocity constraints. To boost the numerical computational efficiency, we introduce a slack variable to reformulate the problem into second-order-cone programming (SOCP). Subsequently, comparative experiments are conducted to validate the superiority of the proposed method. Quantitative performance indexes show that TOPP-DWR achieves TOPP while adhering to all constraints. Finally, field autonomous navigation experiments are carried out to validate the practicability of TOPP-DWR in real-world applications.

EAAI Journal 2024 Journal Article

A simple rapid sample-based clustering for large-scale data

• Yewang Chen
• Yuanyuan Yang
• Songwen Pei
• Yi Chen
• Jixiang Du

Large-scale data clustering is a crucial task in addressing big data challenges. However, existing approaches often struggle to efficiently and effectively identify different types of big data, making it a significant challenge. In this paper, we propose a novel sample-based clustering algorithm, which is very simple but extremely efficient, and runs in about O ( n × r ) expected time, where n is the size of the dataset and r is the category number. The method is based on two key assumptions: (1) The data of each sufficient sample should have similar data distribution, as well as category distribution, to the entire data set; (2) the representative of each category in all sufficient samples conform to Gaussian distribution. It processes data in two stages, one is to classify data in each local sample independently, and the other is to globally classify data by assigning each point to the category of its nearest representative category center. The experimental results show that the proposed algorithm is effective, which outperforms other current variants of clustering algorithm.

ICML Conference 2024 Conference Paper

Constrained Ensemble Exploration for Unsupervised Skill Discovery

• Chenjia Bai
• Rushuai Yang
• Qiaosheng Zhang 0002
• Kang Xu
• Yi Chen
• Ting Xiao 0002
• Xuelong Li 0001

Unsupervised Reinforcement Learning (RL) provides a promising paradigm for learning useful behaviors via reward-free per-training. Existing methods for unsupervised RL mainly conduct empowerment-driven skill discovery or entropy-based exploration. However, empowerment often leads to static skills, and pure exploration only maximizes the state coverage rather than learning useful behaviors. In this paper, we propose a novel unsupervised RL framework via an ensemble of skills, where each skill performs partition exploration based on the state prototypes. Thus, each skill can explore the clustered area locally, and the ensemble skills maximize the overall state coverage. We adopt state-distribution constraints for the skill occupancy and the desired cluster for learning distinguishable skills. Theoretical analysis is provided for the state entropy and the resulting skill distributions. Based on extensive experiments on several challenging tasks, we find our method learns well-explored ensemble skills and achieves superior performance in various downstream tasks compared to previous methods.

AIIM Journal 2024 Journal Article

Enhanced differential evolution algorithm for feature selection in tuberculous pleural effusion clinical characteristics analysis

• Xinsen Zhou
• Yi Chen
• Wenyong Gui
• Ali Asghar Heidari
• Zhennao Cai
• Mingjing Wang
• Huiling Chen
• Chengye Li

Tuberculous pleural effusion poses a significant threat to human health due to its potential for severe disease and mortality. Without timely treatment, it may lead to fatal consequences. Therefore, early identification and prompt treatment are crucial for preventing problems such as chronic lung disease, respiratory failure, and death. This study proposes an enhanced differential evolution algorithm based on colony predation and dispersed foraging strategies. A series of experiments conducted on the IEEE CEC 2017 competition dataset validated the global optimization capability of the method. Additionally, a binary version of the algorithm is introduced to assess the algorithm's ability to address feature selection problems. Comprehensive comparisons of the effectiveness of the proposed algorithm with 8 similar algorithms were conducted using public datasets with feature sizes ranging from 10 to 10, 000. Experimental results demonstrate that the proposed method is an effective feature selection approach. Furthermore, a predictive model for tuberculous pleural effusion is established by integrating the proposed algorithm with support vector machines. The performance of the proposed model is validated using clinical records collected from 140 tuberculous pleural effusion patients, totaling 10, 780 instances. Experimental results indicate that the proposed model can identify key correlated indicators such as pleural effusion adenosine deaminase, temperature, white blood cell count, and pleural effusion color, aiding in the clinical feature analysis of tuberculous pleural effusion and providing early warning for its treatment and prediction.

AAAI Conference 2024 Short Paper

MapLE: Matching Molecular Analogues Promptly with Low Computational Resources by Multi-Metrics Evaluation (Student Abstract)

• Xiaojian Chen
• Chuyue Liao
• Yanhui Gu
• Yafei Li
• Jinlan Wang
• Yi Chen
• Masaru Kitsuregawa

Matching molecular analogues is a computational chemistry and bioinformatics research issue which is used to identify molecules that are structurally or functionally similar to a target molecule. Recent studies on matching analogous molecules have predominantly concentrated on enhancing effectiveness, often sidelining computational efficiency, particularly in contexts of low computational resources. This oversight poses challenges in many real applications (e.g., drug discovery, catalyst generation and so forth). To tackle this issue, we propose a general strategy named MapLE, aiming to promptly match analogous molecules with low computational resources by multi-metrics evaluation. Experimental evaluation conducted on a public biomolecular dataset validates the excellent and efficient performance of the proposed strategy.

NeurIPS Conference 2024 Conference Paper

Pearls from Pebbles: Improved Confidence Functions for Auto-labeling

• Harit Vishwakarma
• Yi Chen
• Sui Jiet Tay
• Satya Sai Srinath Namburi
• Frederic Sala
• Ramya Korlakai Vinayak

Auto-labeling is an important family of techniques that produce labeled training sets with minimum manual annotation. A prominent variant, threshold-based auto-labeling (TBAL), works by finding thresholds on a model's confidence scores above which it can accurately automatically label unlabeled data. However, many models are known to produce overconfident scores, leading to poor TBAL performance. While a natural idea is to apply off-the-shelf calibration methods to alleviate the overconfidence issue, we show that such methods fall short. Rather than experimenting with ad-hoc choices of confidence functions, we propose a framework for studying the optimal TBAL confidence function. We develop a tractable version of the framework to obtain Colander (Confidence functions for Efficient and Reliable Auto-labeling), a new post-hoc method specifically designed to maximize performance in TBAL systems. We perform an extensive empirical evaluation of Colander and compare it against methods designed for calibration. Colander achieves up to 60% improvement on coverage over the baselines while maintaining error level below 5% and using the same amount of labeled data.

EAAI Journal 2024 Journal Article

Personal credit default prediction fusion framework based on self-attention and cross-network algorithms

• Di Han
• Wei Guo
• Yi Chen
• Bocheng Wang
• Wenting Li

As the volume of open data from cloud platforms, including consumer, credit, and social data, experiences exponential growth, the problem of data collection for credit and lending has been effectively alleviated. However, this surge in massive data exhibits new characteristics of high dimensionality and imbalance, which makes the value information density of credit features become very sparse, resulting in the inability of existing data processing methods to extract latent information from the data, and the difficulty of prediction models to assign more accurate weights to crucial features. This affects the model’s performance in assessing individual credit defaults. To address these issues, this paper optimizes the data processing process, then introduces the self-attention and cross-network credit default prediction fusion framework (SACN), which incorporates a cross-network and self-attention mechanism. This fusion framework optimizes the credit data feature engineering process, further reducing conflicts among features from multiple sources. Through experimental comparisons using publicly available credit datasets, SACN accurately captures explicit and implicit high-order data feature interactions within credit lending, enhancing the precise and efficient extraction of critical credit information. Its performance in credit default prediction surpasses that of other mainstream predictive models and maintains accuracy and stability across various types of datasets. The source code is publicly available at https: //gitee. com/andyham_andy. ham/sacn-forecasting-framework. git.

YNICL Journal 2024 Journal Article

Right superior frontal gyrus: A potential neuroimaging biomarker for predicting short-term efficacy in schizophrenia

• Yongfeng Yang
• Xueyan Jin
• Yongjiang Xue
• Xue Li
• Yi Chen
• Ning Kang
• Wei Yan
• Peng Li

Antipsychotic drug treatment for schizophrenia (SZ) can alter brain structure and function, but it is unclear if specific regional changes are associated with treatment outcome. Therefore, we examined the effects of antipsychotic drug treatment on regional grey matter (GM) density, white matter (WM) density, and functional connectivity (FC) as well as associations between regional changes and treatment efficacy. SZ patients (n = 163) and health controls (HCs) (n = 131) were examined by structural magnetic resonance imaging (sMRI) at baseline, and a subset of SZ patients (n = 77) were re-examined after 8 weeks of second-generation antipsychotic treatment to assess changes in regional GM and WM density. In addition, 88 SZ patients and 81 HCs were examined by resting-state functional MRI (rs-fMRI) at baseline and the patients were re-examined post-treatment to examine FC changes. The Positive and Negative Syndrome Scale (PANSS) and MATRICS Consensus Cognitive Battery (MCCB) were applied to measure psychiatric symptoms and cognitive impairments in SZ. SZ patients were then stratified into response and non-response groups according to PANSS score change (≥50 % decrease or <50 % decrease, respectively). The GM density of the right cingulate gyrus, WM density of the right superior frontal gyrus (SFG) plus 5 other WM tracts were reduced in the response group compared to the non-response group. The FC values between the right anterior cingulate and paracingulate gyrus and left thalamus were reduced in the entire SZ group (n = 88) after treatment, while FC between the right inferior temporal gyrus (ITG) and right medial superior frontal gyrus (SFGmed) was increased in the response group. There were no significant changes in regional FC among the non-response group after treatment and no correlations with symptom or cognition test scores. These findings suggest that the right SFG is a critical target of antipsychotic drugs and that WM density and FC alterations within this region could be used as potential indicators in predicting the treatment outcome of antipsychotics of SZ.

IROS Conference 2023 Conference Paper

Need for Speed: Fast Correspondence-Free Lidar-Inertial Odometry Using Doppler Velocity

• David J. Yoon
• Keenan Burnett
• Johann Laconte
• Yi Chen
• Heethesh Vhavle
• Sören Kammel
• James Reuther
• Tim D. Barfoot

In this paper, we present a fast, lightweight odometry method that uses the Doppler velocity measurements from a Frequency-Modulated Continuous-Wave (FMCW) lidar without data association. FMCW lidar is a recently emerging technology that enables per-return relative radial velocity measurements via the Doppler effect. Since the Doppler measurement model is linear with respect to the 6-degrees-of-freedom (DOF) vehicle velocity, we can formulate a linear continuous-time estimation problem for the velocity and numerically integrate for the 6-DOF pose estimate afterward. The caveat is that angular velocity is not observable with a single FMCW lidar. We address this limitation by also incorporating the angular velocity measurements from a gyroscope. This results in an extremely efficient odometry method that processes lidar frames at an average wall-clock time of 5. 64ms on a single thread, well below the 10Hz operating rate of the lidar we tested. We show experimental results on real-world driving sequences and compare against state-of-the-art Iterative Closest Point (ICP)-based odometry methods, presenting a compelling tradeoff between accuracy and computation. We also present an algebraic observability study, where we demonstrate in theory that the Doppler measurements from multiple FMCW lidars are capable of observing all 6 degrees of freedom (translational and angular velocity).

ICRA Conference 2023 Conference Paper

On Locally Optimal Redundancy Resolution using the Basis of the Null Space

• Eugenio Monari
• Yi Chen
• Rocco Vertechy

This paper presents two methods for the computation of the null space velocity command in redundant robots. Both these methods resort to the solution of a constrained optimization problem. The first one is a formalization of the traditional Gradient Projection Method (GPM) which guarantees the respect of the joint bounds and a gradual activation/deactivation of the null space command. The second one, called Null Space Basis Optimal Linear Combination Method (NSBM), finds the optimal coefficients of a basis of the null space of the Jacobian, ensuring in turn that the joint bounds are respected and that the null space is activated and deactivated gradually. The two methods are applied to the case study of a welding application in which the null space command must avoid the collision between the robot and an obstacle. The comparison of the results of the case study shows that NSBM performs better than GPM. The proposed algorithms are also tested on a real robotic platform to demonstrate that their computational time is compatible with the real-time requirements of the robot.

AIIM Journal 2022 Journal Article

DI++: A deep learning system for patient condition identification in clinical notes

• Jinhe Shi
• Xiangyu Gao
• William C. Kinsman
• Chenyu Ha
• Guodong Gordon Gao
• Yi Chen

Accurately recording a patient's medical conditions in an EHR system is the basis of effectively documenting patient health status, coding for billing, and supporting data-driven clinical decision making. However, patient conditions are often not fully captured in structured EHR systems, but may be documented in unstructured clinical notes. The challenge is that not all disease mentions in clinical notes actually refer to a patient's conditions. We developed a two-step workflow for identifying patient's conditions from clinical notes: disease mention extraction and disease mention classification. We implemented this workflow in a prototype system, DI++, for Disease Identification. An advanced deep learning model, CLSTM-Attention model, is developed for disease mention classification in DI++. Extensive empirical evaluation on about one million pages of de-identified clinical notes demonstrates that DI++ has significant performance advantage over existing systems on F1 Score, Area Under the Curve metrics, and efficiency. The proposed CLSTM-Attention model outperforms the existing deep learning models for disease mention classification.

YNIMG Journal 2022 Journal Article

Focal fMRI signal enhancement with implantable inductively coupled detectors

• Yi Chen
• Qi Wang
• Sangcheon Choi
• Hang Zeng
• Kengo Takahashi
• Chunqi Qian
• Xin Yu

Despite extensive efforts to increase the signal-to-noise ratio (SNR) of fMRI images for brain-wide mapping, technical advances of focal brain signal enhancement are lacking, in particular, for animal brain imaging. Emerging studies have combined fMRI with fiber optic-based optogenetics to decipher circuit-specific neuromodulation from meso to macroscales. High-resolution fMRI is needed to integrate hemodynamic responses into cross-scale functional dynamics, but the SNR remains a limiting factor given the complex implantation setup of animal brains. Here, we developed a multimodal fMRI imaging platform with an implanted inductive coil detector. This detector boosts the tSNR of MRI images, showing a 2-3-fold sensitivity gain over conventional coil configuration. In contrast to the cryoprobe or array coils with limited spaces for implanted brain interface, this setup offers a unique advantage to study brain circuit connectivity with optogenetic stimulation and can be further extended to other multimodal fMRI mapping schemes.

IJCAI Conference 2022 Conference Paper

Signed Neuron with Memory: Towards Simple, Accurate and High-Efficient ANN-SNN Conversion

• Yuchen Wang
• Malu Zhang
• Yi Chen
• Hong Qu

Spiking Neural Networks (SNNs) are receiving increasing attention due to their biological plausibility and the potential for ultra-low-power event-driven neuromorphic hardware implementation. Due to the complex temporal dynamics and discontinuity of spikes, training SNNs directly usually suffers from high computing resources and a long training time. As an alternative, SNN can be converted from a pre-trained artificial neural network (ANN) to bypass the difficulty in SNNs learning. However, the existing ANN-to-SNN methods neglect the inconsistency of information transmission between synchronous ANNs and asynchronous SNNs. In this work, we first analyze how the asynchronous spikes in SNNs may cause conversion errors between ANN and SNN. To address this problem, we propose a signed neuron with memory function, which enables almost no accuracy loss during the conversion process, and maintains the properties of asynchronous transmission in the converted SNNs. We further propose a new normalization method, named neuron-wise normalization, to significantly shorten the inference latency in the converted SNNs. We conduct experiments on challenging datasets including CIFAR10 (95. 44% top-1), CIFAR100 (78. 3% top-1) and ImageNet (73. 16% top-1). Experimental results demonstrate that the proposed method outperforms the state-of-the-art works in terms of accuracy and inference time. The code is available at https: //github. com/ppppps/ANN2SNNConversion_SNM_NeuronNorm.

AAAI Conference 2021 Conference Paper

Deep Spiking Neural Network with Neural Oscillation and Spike-Phase Information

• Yi Chen
• Hong Qu
• Malu Zhang
• Yuchen Wang

Deep spiking neural network (DSNN) is a promising computational model towards artificial intelligence. It benefits from both the DNNs and SNNs through a hierarchy structure to extract multiple levels of abstraction and the event-driven computational manner to provide ultra-low-power neuromorphic implementation, respectively. However, how to efficiently train the DSNNs remains an open question because of the non-differentiable spike function that prevents the traditional back-propagation (BP) learning algorithm directly applied to DSNNs. Here, inspired by the findings from the biological neural networks, we address the above-mentioned problem by introducing neural oscillation and spike-phase information to DSNNs. Specifically, we propose an Oscillation Postsynaptic Potential (Os-PSP) and phase-locking active function, and further put forward a new spiking neuron model, namely Resonate Spiking Neuron (RSN). Based on the RSN, we propose a Spike-Level-Dependent Back-Propagation (SLDBP) learning algorithm for DSNNs. Experimental results show that the proposed learning algorithm resolves the problems caused by the incompatibility between the BP learning algorithm and SNNs, and achieves state-of-the-art performance in single spike-based learning algorithms. This work investigates the contribution of introducing biologically inspired mechanisms, such as neural oscillation and spike-phase information to DSNNs and providing a new perspective to design future DSNNs.

TCS Journal 2021 Journal Article

On finding maximum disjoint paths with different colors: Computational complexity and practical LP-based algorithms

• Yunyun Deng
• Longkun Guo
• Kewen Liao
• Yi Chen

With the rapid development of wireless networks, the burden on data transmission is becoming much higher, so are the requirements for bandwidth and load balancing. To cope with these changing requirements, we investigate a novel problem of finding maximum disjoint paths with different colors (MDPDC). In MDPDC, transmission frequencies in a network are modeled as different colors on network nodes. The aim is to find a maximum number of color-constrained node-disjoint paths where nodes must share the same color within any disjoint path, and differ in color among different disjoint paths. For this proposed problem, we first prove MDPDC is NP -complete in both directed and undirected graphs. Then we provide two practical linear programming based solutions with theoretical justifications of their correctness and time complexity. Extensive computer experiments are also carried out with several compared baseline methods to demonstrate the effectiveness of proposed algorithms both in running time and solution quality.

YNIMG Journal 2021 Journal Article

To pool or not to pool: Can we ignore cross-trial variability in FMRI?

• Gang Chen
• Srikanth Padmala
• Yi Chen
• Paul A. Taylor
• Robert W. Cox
• Luiz Pessoa

In this work, we investigate the importance of explicitly accounting for cross-trial variability in neuroimaging data analysis. To attempt to obtain reliable estimates in a task-based experiment, each condition is usually repeated across many trials. The investigator may be interested in (a) condition-level effects, (b) trial-level effects, or (c) the association of trial-level effects with the corresponding behavior data. The typical strategy for condition-level modeling is to create one regressor per condition at the subject level with the underlying assumption that responses do not change across trials. In this methodology of complete pooling, all cross-trial variability is ignored and dismissed as random noise that is swept under the rug of model residuals. Unfortunately, this framework invalidates the generalizability from the confine of specific trials (e.g., particular faces) to the associated stimulus category ("face"), and may inflate the statistical evidence when the trial sample size is not large enough. Here we propose an adaptive and computationally tractable framework that meshes well with the current two-level pipeline and explicitly accounts for trial-by-trial variability. The trial-level effects are first estimated per subject through no pooling. To allow generalizing beyond the particular stimulus set employed, the cross-trial variability is modeled at the population level through partial pooling in a multilevel model, which permits accurate effect estimation and characterization. Alternatively, trial-level estimates can be used to investigate, for example, brain-behavior associations or correlations between brain regions. Furthermore, our approach allows appropriate accounting for serial correlation, handling outliers, adapting to data skew, and capturing nonlinear brain-behavior relationships. By applying a Bayesian multilevel model framework at the level of regions of interest to an experimental dataset, we show how multiple testing can be addressed and full results reported without arbitrary dichotomization. Our approach revealed important differences compared to the conventional method at the condition level, including how the latter can distort effect magnitude and precision. Notably, in some cases our approach led to increased statistical sensitivity. In summary, our proposed framework provides an effective strategy to capture trial-by-trial responses that should be of interest to a wide community of experimentalists.

ICRA Conference 2021 Conference Paper

When Shall I Be Empathetic? The Utility of Empathetic Parameter Estimation in Multi-Agent Interactions

• Yi Chen
• Lei Zhang
• Tanner Merry
• Sunny Amatya
• Wenlong Zhang
• Yi Ren

Human-robot interactions (HRI) can be modeled as differential games with incomplete information, where each agent holds private reward parameters. Due to the open challenge in finding perfect Bayesian equilibria of such games, existing studies often decouple the belief and physical dynamics by iterating between belief update and motion planning. Importantly, the robot’s reward parameters are often assumed to be known to the humans, in order to simplify the computation. We show in this paper that under this simplification, the robot performs non-empathetic belief update about the humans’ parameters, which causes high safety risks in uncontrolled intersection scenarios. In contrast, we propose a model for empathetic belief update, where the agent updates the joint probabilities of all agents’ parameter combinations. The update uses a neural network that approximates the Nash equilibrial action-values of agents. We compare empathetic and non-empathetic belief update methods on a two-vehicle uncontrolled intersection case with short reaction time. Results show that when both agents are unknowingly aggressive (or non-aggressive), empathy is necessary for avoiding collisions when agents have false believes about each others’ parameters. This paper demonstrates the importance of acknowledging the incomplete-information nature of HRI.

IROS Conference 2020 Conference Paper

Enabling Robot to Assist Human in Collaborative Assembly using Convolutional Neural Networks

• Yi Chen
• Weitian Wang
• Venkat Krovi
• Yunyi Jia

Human-robot collaborative assembly consists of humans and automated robots, who cooperate with each other to accomplish complex assembly tasks, which are difficult for either humans or robots to accomplish alone. There has been some success in statistics-based and optimization-based approaches to realize human-robot collaboration. However, they usually need a set of complex modeling and setup efforts and the robots usually need to be programmed by a well-trained expert. In this paper, we take a new approach by introducing convolutional neural networks (CNN) into the teaching- learning-collaboration (TLC) model for collaborative assembly tasks. The proposed approach can alleviate the need for complex modeling and setup compared to the existing approaches. It can collect and automatically label the data from human demonstrations and then train a CNN-based robot assistance model to make the robot assist humans in the assembly process in real-time. We have experimentally verified our proposed approach on a human-robot collaborative assembly platform and the results suggest that the robot can successfully learn from human demonstrations to automatically generate right actions to assist human in accomplishing assembly tasks.

ICRA Conference 2020 Conference Paper

Real-Time Adaptive Assembly Scheduling in Human-Multi-Robot Collaboration According to Human Capability *

• Shaobo Zhang
• Yi Chen
• Jun Zhang
• Yunyi Jia

Human-multi-robot collaboration is becoming more and more common in intelligent manufacturing. Optimal assembly scheduling of such systems plays a critical role in their production efficiency. Existing approaches mostly consider humans as agents with assumed or known capabilities, which leads to suboptimal performance in realistic applications where human capabilities usually change. In addition, most robot adaptation focuses on human-single-robot interaction and the adaptation in human-multi-robot interaction with changing human capability still remains challenging due to the complexity of the heterogeneous multi-agent interactions. This paper proposes a real-time adaptive assembly scheduling approach for human-multi-robot collaboration by modeling and incorporating changing human capability. A genetic algorithm is also designed to derive implementable solutions for the formulated adaptive assembly scheduling problem. The proposed approaches are validated through different simulated human-multi-robot assembly tasks and the results demonstrate the effectiveness and advantages of the proposed approaches.

JBHI Journal 2019 Journal Article

Improved False Positive Reduction by Novel Morphological Features for Computer-Aided Polyp Detection in CT Colonography

• Yacheng Ren
• Jingchen Ma
• Junfeng Xiong
• Yi Chen
• Lin Lu
• Jun Zhao

Computer-aided detection (CAD) systems can assist radiologists in reducing the interpretation time and improving the detection results in computed tomographic colonography (CTC). However, existing false positives (FPs) impair the advantages of CAD systems. This study aims to develop new morphological features for the FP reduction while maintaining high detection sensitivity. Volumetric feature maps are computed for each polyp candidate by using three-dimensional (3-D) geodesic distance transformation, circular transformation (CcT), and quantized convergence index (QCI) filters. Then, new morphological features are developed based on the curvature, fractal dimension, and volumetric feature maps. To the best of our knowledge, we are also the first to develop 3-D CcT and QCI filters specifically for colonic polyps. The new morphological features were evaluated to reduce the FPs by using 456 oral contrast-enhanced CT scans from 228 patients with 130 polyps ≥5 mm. For comparison, the well-defined features from our previous work were used to generate a baseline reference. The additional use of the new morphological features reduced the FP rate from 4. 2 to 2. 0 FPs per scan (i. e. , 52. 4% FP reduction percentage) at 96. 2% by-polyp sensitivity and from 4. 5 to 2. 1 FPs per scan (i. e. , 53. 3% FP reduction percentage) at 93. 9% per-scan sensitivity for polyps ≥5 mm. Experimental results indicate that the new morphological features can effectively reduce the FP rate without sacrificing detection sensitivity. We believe that the newly developed morphological features would advance the CAD systems to assist radiologists in interpreting CTC images.

NeurIPS Conference 2018 Conference Paper

Contrastive Learning from Pairwise Measurements

• Yi Chen
• Zhuoran Yang
• Yuchen Xie
• Zhaoran Wang

Learning from pairwise measurements naturally arises from many applications, such as rank aggregation, ordinal embedding, and crowdsourcing. However, most existing models and algorithms are susceptible to potential model misspecification. In this paper, we study a semiparametric model where the pairwise measurements follow a natural exponential family distribution with an unknown base measure. Such a semiparametric model includes various popular parametric models, such as the Bradley-Terry-Luce model and the paired cardinal model, as special cases. To estimate this semiparametric model without specifying the base measure, we propose a data augmentation technique to create virtual examples, which enables us to define a contrastive estimator. In particular, we prove that such a contrastive estimator is invariant to model misspecification within the natural exponential family, and moreover, attains the optimal statistical rate of convergence up to a logarithmic factor. We provide numerical experiments to corroborate our theory.

JBHI Journal 2018 Journal Article

Robust Activity Recognition for Aging Society

• Yi Chen
• Li Yu
• Kaoru Ota
• Mianxiong Dong

Human activity recognition (HAR) is widely applied to many industrial applications. In the context of Industry 4. 0, driven by the same demand of machines' self-organizing ability, HAR can also be adopted in elderly healthcare. However, HAR should be adaptive to the application scenarios in elderly healthcare. In this paper, we propose a nonintrusive activity recognition method that can be applied to long-term and unobtrusive monitoring for elderlies. The method is robust to obstruction and nontarget object interference. Skeleton sequence is estimated from RGB images. Based on two activity continuity metrics, an interframe matching algorithm is proposed to filter nontarget objects. In order to make full use of spatial-temporal information, we propose a novel activity encoding method based on the interframe joints distances. A convolutional neural network is used to learn the distinguishing features automatically. A specific data augmentation method is designed to avoid the overfitting problem on small-scale datasets. The experiments are performed on two public activity datasets and a newly released noisy activity dataset (NAD). The NAD contains obstruction, nontarget object interference. The experimental results show that the proposed method achieves the state-of-the-art performance while only using one ordinary camera. The proposed method is robust to a realistic environment.

YNIMG Journal 2018 Journal Article

Scale-specific analysis of fMRI data on the irregular cortical surface

• Yi Chen
• Radoslaw Martin Cichy
• Wilhelm Stannat
• John-Dylan Haynes

To fully characterize the activity patterns on the cerebral cortex as measured with fMRI, the spatial scale of the patterns must be ascertained. Here we address this problem by constructing steerable bandpass filters on the discrete, irregular cortical mesh, using an improved Gaussian smoothing in combination with differential operators of directional derivatives. We demonstrate the utility of the algorithm in two ways. First, using modelling we show that our algorithm yields superior results in numerical precision and spatial uniformity of filter kernels compared to the most widely adopted approach for cortical smoothing. As the effective scales of information differ from the nominal filter sizes applied to extract them, we evaluated the effective scales empirically for different filters to make subsequent comparisons well calibrated. Second, we applied the algorithm to an fMRI dataset to assess the scale and pattern form of cortical encoding of information about visual objects in the ventral visual pathway. We found that filtering by our method improved the detection of discriminant information about experimental conditions over previous methods, that the level of categorization (subordinate versus superordinate) of objects was differentially related to the spatial scale of fMRI patterns, and that the spatial scale at which information was encoded increased along the ventral visual pathway. In sum, our results indicate that the proposed algorithm is particularly suited to assess and detect scale-specific information encoding in cortex, and promises further insight into the topography of cortical encoding in the human brain.

AAAI Conference 2017 Conference Paper

Learning Context-Specific Word/Character Embeddings

• Xiaoqing Zheng
• Jiangtao Feng
• Yi Chen
• Haoyuan Peng
• Wenqing Zhang

Unsupervised word representations have demonstrated improvements in predictive generalization on various NLP tasks. Most of the existing models are in fact good at capturing the relatedness among words rather than their “genuine” similarity because the context representations are often represented by a sum (or an average) of the neighbor’s embeddings, which simplifies the computation but ignores an important fact that the meaning of a word is determined by its context, reflecting not only the surrounding words but also the rules used to combine them (i. e. compositionality). On the other hand, much effort has been devoted to learning a singleprototype representation per word, which is problematic because many words are polysemous, and a single-prototype model is incapable of capturing phenomena of homonymy and polysemy. We present a neural network architecture to jointly learn word embeddings and context representations from large data sets. The explicitly produced context representations are further used to learn context-specific and multiprototype word embeddings. Our embeddings were evaluated on several NLP tasks, and the experimental results demonstrated the proposed model outperformed other competitors and is applicable to intrinsically “character-based” languages.

IJCAI Conference 2015 Conference Paper

Character-Based Parsing with Convolutional Neural Network

• Xiaoqing Zheng
• Haoyuan Peng
• Yi Chen
• Pengjing Zhang
• Wenqiang Zhang

We describe a novel convolutional neural network architecture with k-max pooling layer that is able to successfully recover the structure of Chinese sentences. This network can capture active features for unseen segments of a sentence to measure how likely the segments are merged to be the constituents. Given an input sentence, after all the scores of possible segments are computed, an ef- ficient dynamic programming parsing algorithm is used to find the globally optimal parse tree. A similar network is then applied to predict syntactic categories for every node in the parse tree. Our networks archived competitive performance to existing benchmark parsers on the CTB-5 dataset without any task-specific feature engineering.

ICRA Conference 2014 Conference Paper

Modular continuum robotic endoscope design and path planning

• Yi Chen
• Jiahui Liang
• Ian W. Hunter

Robotic endoscopes have the potential to help endoscopists position tools during procedures, to propel the endoscope to the desired position, to automate functions and to prevent perforations during procedures. This paper outlines the modular architecture for a continuum robotic endoscope with multiple bending segments along the length of the endoscope. Each of the segments is modular, containing a set of actuation motors that drive short cables in the continuum segments. Each modular segment of the robot is 15 mm in diameter, can turn 180 degrees and has a turning speed ranging from 35 to 250 degrees per second. The robot is composed of seven of these modular segments, has 14 degrees of freedom, is 0. 91 m long and has a mass of 157 grams. The implementation for the mechanical, electrical, and software design is described and the robotic endoscope bending motions are sensed, simulated and controlled using kinematic models. Lastly, path planning trajectories of the endoscope segments are designed and coordinated to help propel the robot forward in an uncoiling motion and in a follow-the-leader fashion along a path that emulates simplified turns in a colon. We show that the robotic endoscope is able to exert less force on the walls of the colon emulation path, enable automated insertion into the patient, and execute colon wall avoidance and linear scanning motions not available in conventional endoscopes.

YNIMG Journal 2013 Journal Article

Statistical inference and multiple testing correction in classification-based multi-voxel pattern analysis (MVPA): Random permutations and cluster size control

• Johannes Stelzer
• Yi Chen
• Robert Turner

An ever-increasing number of functional magnetic resonance imaging (fMRI) studies are now using information-based multi-voxel pattern analysis (MVPA) techniques to decode mental states. In doing so, they achieve a significantly greater sensitivity compared to when they use univariate frameworks. However, the new brain-decoding methods have also posed new challenges for analysis and statistical inference on the group level. We discuss why the usual procedure of performing t-tests on accuracy maps across subjects in order to produce a group statistic is inappropriate. We propose a solution to this problem for local MVPA approaches, which achieves higher sensitivity than other procedures. Our method uses random permutation tests on the single-subject level, and then combines the results on the group level with a bootstrap method. To preserve the spatial dependency induced by local MVPA methods, we generate a random permutation set and keep it fixed across all locations. This enables us to later apply a cluster size control for the multiple testing problem. More specifically, we explicitly compute the distribution of cluster sizes and use this to determine the p-values for each cluster. Using a volumetric searchlight decoding procedure, we demonstrate the validity and sensitivity of our approach using both simulated and real fMRI data sets. In comparison to the standard t-test procedure implemented in SPM8, our results showed a higher sensitivity. We discuss the theoretical applicability and the practical advantages of our approach, and outline its generalization to other local MVPA methods, such as surface decoding techniques.

ICRA Conference 2013 Conference Paper

Two-axis bend sensor design, kinematics and control for a continuum robotic endoscope

• Yi Chen
• Jillian M. Oliveira
• Ian W. Hunter

Angular displacement sensing and contact force sensing in robotic surgery is important for finding the conformation of the tool and for closed loop control. Current techniques for measuring large deflections, such as electromagnetic tracking, involve bulky external reference equipment. This paper discusses the design, dynamic modeling, and kinematic simulation of a compact two-axis bend sensor constructed from layers of carbon black filled polyurethane for determining the bending angle of a cable-driven continuum robotic endoscope. The sensor is flexible ( 10% strain), compact, and does not require external reference equipment. The sensor is able to bend in 180 degrees in all directions, produces two-axis outputs that are linear up to 120 degrees of bending and has a response time less than 9 ms. This paper explores the use of the sensor for controlling the tip position of a robotic endoscope and detecting external forces. The design outlined in this paper has many possible applications in robotic endoscopes, surgical tools, snake-like robots and soft material robotics.

YNIMG Journal 2011 Journal Article

Brain atrophy correlates with functional outcome in a murine model of multiple sclerosis

• I. Pirko
• A.J. Johnson
• Yi Chen
• D.M. Lindquist
• A.K. Lohrey
• J. Ying
• R. Scott Dunn

White matter (WM) lesions are the classic pathological hallmarks of multiple sclerosis (MS). However, MRI-based WM lesion load shows relatively poor correlation with functional outcome, resulting in the “clinico-radiological paradox” of MS. Unlike lesion based measures, volumetric MRI assessment of brain atrophy shows a strong correlation with functional outcome, and the presence of early atrophy predicts a worse disease course. While extensive literature exists describing MRI characteristics of atrophy in MS, the exact pathogenesis and the substrate of atrophy–gray vs. WM loss, axonal/neuronal damage vs. demyelination, or a combination of the above–remain unclear. Animal models of atrophy would allow for detailed investigations of the pathomechanism, and would contribute to an enhanced understanding of structural–functional connections in this complex disease. We now report that in the Theiler's Murine Encephalitis Virus (TMEV) model of MS in SJL/J mice, significant brain atrophy accompanies the development of the progressive MS-like disease. We conducted volumetric MRI studies in 8 cases and 4 age, gender- and strain-matched control mice. While in controls we did not detect any brain atrophy, significant atrophy developed as early as 3months into the disease course, and reached its peak by 6months, resulting in ventricular enlargement by 118% (p =0. 00003). A strong correlation (r =−0. 88) between atrophy and disability, as assessed by rotarod assay, was also demonstrated. We earlier reported another neurodegenerative feature in this model, the presence of deep gray matter T2 hypointensity in thalamic nuclei. Future studies utilizing this model will allow us to investigate key components of MRI detectable neurodegenerative feature development, their tissue correlations and associations with functional outcome measures. These studies are expected to pave the way to a better understanding of the substrate of disability in MS models.

YNIMG Journal 2011 Journal Article

Cortical surface-based searchlight decoding

• Yi Chen
• Praneeth Namburi
• Lloyd T. Elliott
• Jakob Heinzle
• Chun Siong Soon
• Michael W.L. Chee
• John-Dylan Haynes

Local voxel patterns of fMRI signals contain specific information about cognitive processes ranging from basic sensory processing to high level decision making. These patterns can be detected using multivariate pattern classification, and localization of these patterns can be achieved with searchlight methods in which the information content of spherical sub-volumes of the fMRI signal is assessed. The only assumption made by this approach is that the patterns are spatially local. We present a cortical surface-based searchlight approach to pattern localization. Voxels are grouped according to distance along the cortical surface—the intrinsic metric of cortical anatomy—rather than Euclidean distance as in volumetric searchlights. Using a paradigm in which the category of visually presented objects is decoded, we compare the surface-based method to a standard volumetric searchlight technique. Group analyses of accuracy maps produced by both methods show similar distributions of informative regions. The surface-based method achieves a finer spatial specificity with comparable peak values of significance, while the volumetric method appears to be more sensitive to small informative regions and might also capture information not located directly within the gray matter. Furthermore, our findings show that a surface centered in the middle of the gray matter contains more information than to the white–gray boundary or the pial surface.

YNIMG Journal 2011 Journal Article

Effective connectivity of brain networks during self-initiated movement in Parkinson's disease

• Tao Wu
• Liang Wang
• Mark Hallett
• Yi Chen
• Kuncheng Li
• Piu Chan

Patients with Parkinson's disease (PD) have difficulty in performing self-initiated movements. The neural mechanism of this deficiency remains unclear. In the current study, we used functional MRI (fMRI) and psychophysiological interaction (PPI) methods to investigate the changes in effective connectivity of the brain networks during performance of self-initiated movement in PD patients. Effective connectivity is defined as the influence one neuronal system exerts over another. fMRIs were acquired in 18 PD patients and in 18 age- and sex-matched healthy controls, when performing a self-initiated right hand tapping task. We chose the left primary motor cortex (M1), rostral supplementary motor area (pre-SMA), left premotor cortex (PMC), left putamen, and right cerebellum as index areas for PPI analysis. During the performance of self-initiated movement, connectivity between the putamen and M1, PMC, SMA, and cerebellum was decreased in PD patients compared to controls. In contrast, connections between the M1, pre-SMA, PMC, parietal cortex, and cerebellum were increased in PD patients compared to controls. In addition, the M1, pre-SMA, PMC, and cerebellum also had less connectivity with the dorsal lateral prefrontal cortex in PD. In PD patients, the effective connectivity between the putamen and M1, PMC, SMA, and cerebellum negatively correlated with the Unified Parkinson's Disease Rating Scale (UPDRS) motor scores; whereas the connectivity between the M1, pre-SMA, PMC, and cerebellum positively correlated with the UPDRS motor scores. Our findings demonstrate that the pattern of interactions of brain networks is disrupted in PD during performance of self-initiated movements. The striatum-cortical and striatum-cerebellar connections are weakened. In contrast, the connections between cortico-cerebellar motor regions are strengthened and may compensate for basal ganglia dysfunction. These altered interregional connections are more deviant when the disorder is more severe, and, therefore, our results give further insight into the explanation for the difficulty in performing self-initiated movements in PD.

YNIMG Journal 2011 Journal Article

Encoding the identity and location of objects in human LOC

• Radoslaw Martin Cichy
• Yi Chen
• John-Dylan Haynes

We are able to recognize objects independent of their location in the visual field. At the same time, we also keep track of the location of objects to orient ourselves and to interact with the environment. The lateral occipital complex (LOC) has been suggested as the prime cortical region for representation of object identity. However, the extent to which LOC also represents object location has remained debated. In this study we used high-resolution fMRI in combination with multivoxel pattern classification to investigate the cortical encoding of three object exemplars from four different categories presented in two different locations. This approach allowed us to study location-tolerant object information and object-tolerant location information in LOC, both at the level of categories and exemplars. We found evidence for both location-tolerant object information and object-tolerant location information in LOC at the level of categories and exemplars. Our results further highlight the mixing of identity and location information in the ventral visual pathway.

ICRA Conference 2010 Conference Paper

Multi-turn, tension-stiffening catheter navigation system

• Yi Chen
• Jean H. Chang
• Alison S. Greenlee
• Kenneth C. Cheung
• Alexander H. Slocum
• Rajiv Gupta

In poorly constrained extra-vascular environments such as hollow viscera, current catheter navigation techniques are restricted to simple paths and therefore limit a doctor's ability to position the catheter. This paper presents a new catheter positioning system that enables faster and more accurate catheter placement, with fewer scans. The proposed robotic catheter navigation system can execute curved paths and maintain any number of three-dimensional turns using tension stiffening guide-wires composed of a set of disposable friction-locking beads. An external, reusable control system is used to automate the movement of the catheter. This control system uses a custom-designed graphical kinematic analysis program that predicts contact forces, changes in conformation due to external forces, tip deflection and failure modes of the catheter as it advances.

YNIMG Journal 2009 Journal Article

Use field map to evaluate iron content in the brains of Huntington's disease patients.

• Yi Chen
• SY Lee
• NK Chen
• DH Salat
• KK Kwong
• HD Rosas

TIME Conference 2004 Conference Paper

Max-Count Aggregation Estimation for Moving Points

• Yi Chen
• Peter Z. Revesz

Many interesting problems regarding moving objects can be reduced to the following question: Given a set S of moving points on a line and two other movings points A and B on the same line, what is the maximum number of points in S that will be simultaneously between A and B within a time interval (t/sub 1/, t/sub 2/)? We propose an algorithm that can estimate the answer for arbitrary A and B and any fixed S in a chosen constant time. We show that the error rate of the estimation is related to this chosen constant and some other parameters of the input data. Our experimental results show that high accuracy estimation can be achieved when S has a large number of points and A and B are not too close to each other.

TIME Conference 2003 Conference Paper

Efficient Aggregation over Moving Objects

• Peter Z. Revesz
• Yi Chen

We study two types of aggregation queries over a set S moving point objects. The first asks to count the number of points in S that are dominated by a query point Q at a given time t. The second asks to find the maximum number of points in S that are dominated by a query point at any time. These queries have several applications in the area of Geographic Information Systems and spatiotemporal databases. For the first query and any fixed dimension d, we give two different solutions, one using O (/spl radic/ N) time and O (N) space and another using O (log N) time and O (N/sup 2/ space, where N is the number of moving points. When each of the points in S is moving piecewise linearly along the same line and the total number of pieces is O (N), then we can do the count query in O (/spl radic/ N) time and O (N) space. For the second query, when all objects move along the x-axis, we give a solution that uses O (log N) time and O (N/sup 2/) space in the worst case. Our solutions introduce novel search structures that can have other applications.